Seagate Technology Holdings plc (STX) Q1 2013 Earnings Report, Transcript and Summary

Good afternoon, and welcome to the Seagate Technology Fiscal First Quarter 2013 Financial Results Conference Call. My name is John, and I'll be your coordinator for today. [Operator Instructions] Following the prepared remarks, there will be a question-and-answer session. As a reminder, this conference is being recorded for replay purposes. At this time, I'd like to turn the call over to Kate Scolnick, Vice President, Investor Relations. Please proceed, Kate.

Good morning, and welcome to today's call. I'm joined today in Cupertino by Seagate CEO, Steve Luczo; CFO, Pat O'Malley; Dave Mosley, EVP of Operations; and Rocky Pimentel, EVP of Sales. We have posted our press release and detailed supplemental information about our fiscal first quarter 2013 on our Investor Relations site at seagate.com. During today's call, Steve will review the highlights from the September quarter and provide the company's outlook for the December quarter. After that, we will open up the call to questions. As a reminder, this conference call contains forward-looking statements, included, but not limited to, statements related to the company's historical and currently anticipated future operating and financial performance in the September quarter and thereafter, and includes statements regarding customer demand for disk drives and general market conditions. These forward-looking statements are based on information available to Seagate as of the date of this conference call and are subject to a number of risks, uncertainties and other factors that could cause actual results to differ materially from those anticipated by these forward-looking statements. Such risks and uncertainties such as global economic conditions and other factors may be beyond the company's control and may pose a risk to the company's operating and financial performance. Information concerning additional factors that could cause results to differ materially from those projected in the forward-looking statements is contained in the company's annual report Form 10-K, as filed with the SEC on August 8, 2012. These forward-looking statements should not be relied upon as representing the company's views of any subsequent date, and Seagate undertakes no obligation to update forward-looking statements to reflect events or circumstances after the date they were made. I would now like to turn the call over to Steve Luczo. Please go ahead, Steve.

Thank you, Kate. Good morning, everyone, and thank you for joining us today. I'd first like to express our concern and sympathies to those that have been impacted by the storm on the East Coast and are enduring difficult times. I also want to thank all past and present Seagate employees, as well as local and central government officials in China, as we just celebrated the production of our 1 billionth drive from our China operations. We accomplished this remarkable feat in just 17 years, and we expect that the company will ship its 2 billionth disk drive this quarter. Congratulations to all involved. Seagate's first quarter results reflect strong operational performance as the company responded quickly to demand and product mix shifts in a challenging environment. We effectively adjusted production and maintained inventory turns in our target range and ended the quarter with inventory levels lower than the June quarter. For the September quarter, we achieved revenues of $3.7 billion, non-GAAP net income of $594 million and non-GAAP diluted earnings per share of $1.45. We maintained our commitment to returning value to shareholders with over 70% of the $1.1 billion in operating cash flow generated during the quarter, returned to shareholders through share redemptions and dividends. Over the last 3 quarters, Seagate's ordinary share count has been reduced by 17%, and dilution from options and equity awards has been reduced by 45%. As with most global technology companies, our financial performance is being impacted by a wide range of factors, including: sluggish macroeconomic growth in all major markets, as well as a slowdown in the BRIC economies; business-spending restraint related to the U.S. federal deficits and the upcoming fiscal cliff in the United States; supply chain and systems inventory adjustments by our major PC OEMs; weaker enterprise demand in…

[Operator Instructions] Our first question comes from Rob Cihra from Evercore Partners.

I just want to ask a question on the, I guess, the enterprise weakness or shortfall in the September quarter and then continuing to the December quarter. Maybe Steve or whomever. How much of that do you think is the market because you did cite Europe? And how much of that was sort of inventory drawdown, and where we are in that inventory drawdown process?

Yes, it's the right question. And unfortunately, it's a little hard to see, because I think the third element that's probably playing into that is the architectural changes as well. We're definitely seeing more customers that are building various types of either private or public cloud infrastructures that are purchasing systems maybe away from the traditional OEMs, whether or not they're going directly or whether or not they may be buying architectures from smaller start-ups. But I think it's the right combination between people that acquired inventory in the March and June quarters in anticipation or concern around recovery still in enterprise from the supply chain disruptions caused by the flood combined with the slowdown. So I think there is an inventory overhang that's still being addressed. And then there's also, I think, some competitive issues where certain of the OEMs maybe just aren't having the sell-through that they thought they were, which may be related to macro or may be competitive issues. But it's definitely a combination of the 3 things of a changing customer base that's responding to a new architecture and then inventory absorption, as well as -- I do think the macro impact is significant, though I've been a little surprised that the European situation hasn't kind of shown its head earlier just because most of the enterprise companies still have somewhere between 25% and 40% of the revenue exposure in Europe, which is different than the client side or consumer technology companies.

Okay. And I guess just, everything you just said, do you think that applies equally to mission critical and business critical, or is most of that more mission critical?

I think it applies equally. I think, probably, the comments around cloud service providers and architectures is probably more of a nearline business critical point than mission critical, but I think there basically was inventory buy-in on nearline as well. Again, a lot of the OEMs feed the cloud market as well, right? So I think it applies to both of them.

Our next question comes from Keith Bachman from Bank of Montréal.

Steve, just want to try to get your take about the OEM inventory. I know you talked a little bit about inventory more broadly, but specifically at the OEM in the channel? And then I had a follow-up please.

Well, I think there's 2 levels of inventory, obviously, at the OEMs, right? One is the device level inventory and then the other is the system level inventory. And I think the comments I just made, I think, at the device level, I think there was probably a fair amount of buy-in really in the June quarter with still some concerns around the recovery of the drive industry. And then I think that, that buy-in happened to coincide with the slowdown that really started at the tail end of the June quarter. I mean to me, 2012 feels a lot like 2010, where we start getting this rollover in May and then things never recovered for the rest of the year. And I think for the OEMs, they were buying in inventory, and then things started to slow down. And it's always a little confusing in June when things slow down because it's the summer, and that's a typically slow period anyhow. And you hope it recovers in September. We really didn't see that recovery, and I think people, clearly, are grasping the macroeconomic issues a little better now as being more significant than maybe we thought 3, 4, 5 months ago. So I think at the OEM level, it's a combination of a lot of system inventory, and initially, in the summer, and device inventory. Then I think probably, trying to bleed down that inventory in anticipation of Windows 8. Clearly, people are trying to get lean on how many of the old systems they're carrying in order to be prepared to ramp as Windows 8's released. And then depending on the robustness of demand for Windows 8, I think they'll be able to respond. So I think it's -- again, I think it's a similar story, just different reasons, maybe more related to Windows as opposed to cloud architecture changing.

But in terms of just run rates, do you feel like their finished goods inventory and/or channel is at a point where the ships of your drives will match whatever their end demand is? So do you think things are kind of evened out? I know it's hard to see.

Well, I think the channel actually feels better to me, if you're talking about -- the distribution channel, with respect to device level technology, feels better. Our position is about 1 million units less than it was last quarter and within our range of 4 to 6 weeks. I'm not that concerned about the distribution channel. I think in the OEMs, there's -- depending on the customer, there's probably still some work to be done. I would say at the margin, I feel -- I wish inventories at the customer level were a little lower, both on the systems side probably and the device side. So again, that's one of the reasons that's weighing on our flat TAM assumption. I think probably end user TAM is actually probably going to grow by low-single digits or maybe even mid-single digits. But I still think there's inventory adjustments to be had.

Okay. My follow-up, then I'll cede the floor is -- Steve, you mentioned you thought ASPs in the March quarter would be flat. I just wanted to see if you could flush that out a little bit. It would be unusual relative to history for those March quarter ASPs to be flattish. Just curious, a little bit of feedback.

Thank you. Well, yes, the difference in history, I suppose, is the structure of the industry. I think the major point that we're trying to convey is in -- this quarter is when we negotiate prices for next quarter. So if you think about it in reverse, for the quarter we're in now, we negotiated those prices in September or late August. And I would say that I kind of feel that the December pricing probably got away from us just a touch with the demand falling off as much as it did in September. And trying to, I think, recognize what was the real TAM going to be for the December quarter, I think the industry and probably our customers were hoping that it was going to be more in the 150 or 155 range. So I think the industry is probably recognized that the March quarter is likely to be pretty flat to or maybe marginally up. And that we're really not going to buy TAM with pricing. And so I think, we've obviously gotten to pricing levels that are as low as we want to see them if we're bumping on the bottom end of our range. And so, it's important for us to make sure we maintain discipline going forward, so we can get our R&D and capital investments where we want for the long term. So that's the way we're going to approach the March quarter.

Our next question comes from Steven Fox from Cross Research.

A couple of questions. First of all, just further on the March pricing and looking forward. Can you sort of describe what kind of mix you anticipate in March, how much that's helping the overall ASP? And then secondly, Steve, you talked a lot about macro pressures that are hard to figure out when they're going to end and so on. If this goes on, say, for 9, 12 months, is there plans to take capacity down? Or do you think you're comfortable operating in these margins on the capacity level as you've got currently?

Thanks, let me just make some notes, so I don't forget. On the first thing, on March pricing, I think, Steven, the mix assumptions -- look, we've been kind of off on mix for a couple of 3 quarters in a row here where the mix assumptions we've had, basically, don't come to fruition. And part, I think, is that's just because the pressure that the systems vendors are under to maintain margins. I think mix is -- the assumptions for mix are basically pretty conservative for the March quarter. So if there was actually an improvement in mix, which would be what we would hope, especially if there's any sort of recovery in March on the enterprise, then I think that the relatively flat ASPs, hopefully, will prove to be conservative. So we have tried to, at least this time, assume a pretty conservative view on mix, to say relatively flat ASPs, because as you know, obviously, with new products coming on and mix improving, it actually probably increase ASPs. On the macroeconomic situation and how long does it have to last for it to be really painful. That's a good question. And I would say if we had -- I guess, the question is do you have clarity on that it's going to be 9 to 12 months. If we knew it was going to be flat and weak for the next 9 to 12 months, we'd probably do one set of things. But if you're kind of making that decision quarter-to-quarter, you have a little bit less flexibility, obviously. But to answer your question, I think the good news is for the drive industry, or certainly, for Seagate, and I guess getting to be that way for WD and Hitachi, is for us, it's about absorbing heads and disks really. It's where most of our R&D and where most of our capital is deployed. And as long as we're absorbing heads and disks at these levels, we're okay. And if we had to reduce drive output, that's actually fairly easy capital for us to address, in part, because a lot of it is test capacity, which we can actually increase test times and also with aerial density -- sorry, with average capacity per drive increasing, test times are increasing anyhow. So in a certain sense, as challenging as the environment is, I'd say for the drive industry, given the aerial density and the petabyte growth, it's manageable. Now if TAMs dropped to 120 million units, that's a different story. Then the industry would probably have to take a hard look at a lot of things with respect to internal components and supply, as well as drive level capacity. But I'd say right now for us, while we're operating well below our theoretical capacity, we can maintain the margin structures and the cost objectives that we have at this level of demand, even if it lasted 9 or 12 months.

Our next question comes from Ananda Baruah from Brean Capital. Ananda Baruah - Brean Murray, Carret & Co., LLC, Research Division: Steve, two if I could. I guess the first one is regarding the gross margin and ASP guidance in December quarter. I guess I was thinking, if prices were coming down about 5%, that gross margins would be able to be stable, more or less. I mean is that just -- sort of typically, not the right sort of assumption? Or is there something going on with mix that's baked into that -- to the gross margin guidance as well? And then I have a follow-up.

Yes, like I said, I think we've been pretty conservative on mix. But again, price erosion is a tricky little term in the drive industry that's, unfortunately, got a legacy that went back to when there was only about 5 different products as opposed to 700 different products. So we like to talk more in terms of ASPs. So you're generally right though that if it was a 5% price erosion that the ASPs probably wouldn't erode by 5%, and gross margins would be healthier. So price erosion is higher than 5%, ASP erosion will be about 5%. And if mix improves either within each category, right -- mix meaning capacity per drive, if you will, or heads and disks per drive, or if the mix improves in terms of enterprise nearline versus client, notebook and desktop, those are all things obviously that boost ASPs. We've taken a pretty conservative approach, as I said on the enterprise market opportunity, in part because of the inventory and in part because of European concerns. Ananda Baruah - Brean Murray, Carret & Co., LLC, Research Division: And just as a follow-up, you've been pretty consistent or very consistent, I should say, on capital allocation through this calendar year. Can you give us some sense of what capital allocation could look like as we get into '13? I believe you have some stipulations around some NOLs. So was interested in sort of getting your thoughts on how that might impact any of the capital allocation.

It doesn't change the long-term capital allocations, where we still believe that we're going to target a heavy return of cash to our shareholders in the form of dividends or redemption. And of course, we've got the 2014 end of year target of $250 million ordinary shares outstanding. So what happens in 2013 though, as you pointed out, is we do run into a Section 382 problem where, if we redeem too many shares, it could trigger a loss of our NOLs, which are fairly substantial. And we'd rather avoid doing that, not that we've used NOLs or necessarily see any need to use NOLs. But it's nice to have them in case you need them. So we'll probably be reasonable in sidestepping that problem, unless the value proposition is just too great, in which case, obviously then, we'll go ahead and buy our stock and deal with losing NOLs that we may or may not need. So what that would mean obviously is a slowdown in redemption in 2013 and then an acceleration in 2014 to get to our target. And in the meantime, obviously, maintaining our dividend policy, which is, as we said, a minimum increase, targeted increase of 10% per year. Ananda Baruah - Brean Murray, Carret & Co., LLC, Research Division: Got it. Is there any sort of level that we should expect the redemptions to sort of settle in at as we get into '13?

They're pretty minimal for '13. Patrick J. O’Malley: It's really for 9 months of '13. In the December quarter of 2013, some of the limitations ease. But for the remainder, the first 9 months of calendar 2013, they will be pretty minimal.

The only thing that can also change that is there are certain holders of securities that we can buy from and not trigger a 382 problem, if you will. Not to imply that anything our government does to us is a problem. But one of those would be Samsung. So if Samsung decided to sell any shares, we could acquire from Samsung and not run into a 382 problem. There are also a couple of institutional holders that we can buy shares from and also not trigger it. So we have a couple other options, but we'll just have to see how it unfolds.

Our next question comes from Sherri Scribner from Deutsche Bank.

I appreciate the detail on the hybrid drive progress. I know you mentioned the 5-millimeter and the 7-millimeter being in the process of being qualified with some OEMs. Can you give us an update on where you are with hybrids? Have you shipped any into new products, or is this really more of a 2013 thing?

It's calendar 2013. So I think by the end of this calendar year, we're going to have hybrid offerings; CTU level, at least, with a lot of major customers, so that would, Sherri, include desktop and enterprise. The 7-millimeter hybrid's getting a lot of attention, which would be a drive that would be used in thin and light. 5-millimeter is not a hybrid, the first version, that would be later. But we think there's lots of interesting applications for 5-millimeter regardless. And I'd say where I'm most excited is the attention that the hybrid in enterprise is getting at a couple of accounts, because you basically can boost IOPS performance by 2x to 3x, which in a lot of applications, IOPS is the measure that they're most concerned about. So we have a couple of large enterprise customers that are very interested in the product, and that's probably the one I'm most excited about. So that's all stuff that we would expect to be ramping in calendar 2013, but some of those even in the first half.

Okay, and then just a quick question on share count. I think the share count was a bit different this quarter than, at least, we had expected. What is the assumption for share count in the December quarter? I know you've given a 350 million by the end of the year, but what type of average should we use for the quarter? Patrick J. O’Malley: I'd use 390 million shares. That may be slightly conservative, depending on the pace we buy to get to 350, but I'd plan for that, Sherri.

Our next question comes from Aaron Rakers from Stifel, Nicolaus. Aaron C. Rakers - Stifel, Nicolaus & Co., Inc., Research Division: Two if I can. I just want go back to the March discussion. Understanding that you are going to be more disciplined relative to December on pricing, is your assumption also, as we work through this OEM inventory that -- what's your assumption around the industry demand TAM or the shipment numbers going into the March quarter? Are we to assume a flat sequential TAM, or do we expect -- I think it's hard to try and judge what normal seasonality is anymore?

Yes, I mean flat. I mean people can get all hung up on 10 million units these days. But the reality is, 10 million units, when you think about our 42% market share, is 4 million units, which I think we could probably respond to in about 1.5 days, so it's not a big deal, right? Whether or not I say 140 or 150, while some analysts may get all exercised about that, the reality is it's the same number from an operations perspective. So I would assume a flat TAM, whether or not that means 140 or even 155, I think is possible. Like I said, I think the December system sales will probably actually be up. My guess is 3% to 7%, but I think we're not going to see it just because of inventory absorption. So that may mean real TAM for December, in terms of sales out of systems in the 150 range, in which case, then March probably, for us, sales will be like that. So that's what we're assuming, but it doesn't really matter. The point is, capacity is way above that. Capacity is 170, so you're not going to pick up anything. You're not going to pick up 30 million units with pricing. The issue is not pricing of systems. The issue is macroeconomic conditions. So the TAM's 140. We'll build whatever it is, 56 million drives, and we're going to price them so we can maintain the margins that we need to invest in our business long term. Aaron C. Rakers - Stifel, Nicolaus & Co., Inc., Research Division: Very helpful. And as a follow-up, you guys had the Analyst Day on September 21. You talked about roughly a 30% gross margin. I'd like to understand, I know it's just a week or so left in the quarter, what changed relative to that? I think you also had made a comment about Brazil having 100 basis points impact. I don't know if that was really the driver. I'd like to understand why that changed in the final week or so of the quarter.

Sure. I don't know if you had heard the earlier comments that I made, but the quarter was really back-end loaded, much more so than -- more so than even is typical for a September. And September's a pretty back-end loaded quarter anyhow as you know. July is usually very weak, August is weak, almost through the entire month. And then it's this race to the finish, and it was even more seasonally back-end loaded. So we had a lot of drives to ship. We looked at historical seasonality, and basically, the quarter didn't come through. So part of it was just related to the overall weakness in demand that basically never really kind of resolved itself. Aaron C. Rakers - Stifel, Nicolaus & Co., Inc., Research Division: The Brazil issue?

So part of it was mix. Again, enterprise mix at the end didn't come in where we thought it would. So again, we have a lot of historical data about what happens with enterprise shipments through the quarter. They're typically even more back-end loaded than client. And so they didn't resolve as much, and the mix wasn't as good. So that was the second thing. And then you have Brazil was like 100 basis points. So I mean even with the first 2 items, we would have hit the 30%, if it wasn't for the Brazil issue. And that was basically a function of -- it's a fairly complicated story. But in Brazil, there is a requirement that if you are building computer systems in Brazil, a certain number or a certain percentage of the components have to be purchased in-country. And we've invested a fair amount of money, in part, as an extension of the investments that Samsung had to make sure that we can actually provide in-country-built drives. Our estimation of what that business should have been in the last couple of quarters was high, and we're still digging into why that is. But as a result, the pricing that we deployed was too aggressive because we didn't get the absorption that we should have. So we've made the adjustments in terms of our expectations. We'll make adjustments in terms of pricing. And we won't get all of that 100 basis points back next quarter, but we think -- this quarter, but we'll probably get half of it back, and then we should kind of be through that knothole [ph].

Our next question comes from Ben Reitzes from Barclays.

I was wondering 2 things. You said OpEx is going to be up sequentially slightly. I was wondering how you were thinking about that lever. I think you mentioned that you didn't want to cut product development, but if things were to get particularly difficult, what are you thinking about the OpEx lever, and what could you do? And I have a follow-up.

Well, we could do a lot on OpEx, as you know, Ben. I think the only thing that's really driving -- and that would again be if we had a forecast of a sustained environment that's a lot worse than this. Again, I'm not going to cut OpEx when we're running at margin levels, like I said, that are higher than what the high end of our old range was. If this is what bad is, we can live with it and continue to invest. The biggest thing driving the increase in the OpEx right now is we hadn't done a focal increase really for a couple of years. And so we're doing a focal increase on comp. And the second is LaCie acquisition brings some OpEx into the model that wasn't there previously. So we know how to manage OpEx if things get bad and stay bad, but I don't think we're near that right now. But we do planning all the time.

My follow-up is on enterprise. Within your TAM estimate, I think you're saying there's continued pressure into the fourth quarter. And just within enterprise, if you wouldn't mind, what are you seeing into December and then into March in terms of TAM? And how we balance the inventories there, just in enterprise, in particular?

Yes, again, we're maintaining a pretty conservative outlook in December and March. Our issue around enterprise, or I think, the technology industry's issue around enterprise is, in my opinion, more related to European situation and weakness in economy just because Europe plays a big role in enterprise purchases, as do governments. And there's obviously more and more policies around austerity, which ultimately means fewer purchases by government. So I think we're just being conservative, there's not going to be much of a recovery, Ben, for a couple of quarters.

So it's like in line with your TAM estimate, just flat and flat kind of is best guess for now?

Yes, that's right.

Our next question comes from Jayson Noland from Robert Baird. Jayson Noland - Robert W. Baird & Co. Incorporated, Research Division: I wanted to ask about, Steve, your comment on balancing supply and demand, strong inventory management and gross margin performance relative to pre-consolidation, especially considering linearity. If you can talk about balancing supply with these conditions, it just seems like Seagate and the industry are more nimble today.

Yes, well, I think Seagate is. I mean if you look at our inventory position, I think it's -- to have less inventory than we did in June, in a quarter that fell off pretty dramatically relative to expectations, I feel that the company responded pretty well. I think linearity is a different issue. I think -- I still think that we and our customers have a lot of work to do around linearity and velocity. And Dave Mosley can talk just a little bit, with Rocky. I mean we're doing a lot of -- it's funny. The flood, for all of its challenges, also, I think, resulted in some really good business process improvements between us and our customers in order to reduce inventory and increase velocity. Some customers are continuing with that. Others are going back to their lazy ways that basically somehow revolve around carrying a lot of inventory to solve business process problems. And I just don't think that's healthy. I think our industry's got to address that with our customers that at the end of the day, that's not good for anybody. So the only way we can really do that is keeping a really tight control on builds. Dave's done a lot of work in this area on the supply chain side, all the way through demand fulfillment. So maybe I'll just ask him to talk to you a little bit, and then Rocky can weigh in on what he's seeing on the customer side in terms of those that are kind of moving towards a model that is maybe more focused on velocity versus inventory.

First of all, product simplification happens, which we've been working really hard, especially through the product transitions that we went through. That helps it straight out of the factories. And then working with our customers to develop hubs that make sense versus hubs that are spread all over the place where we can late-stage postpone some of the configuration issues that they need out the field. With product simplification, that's helping us quite a bit. And then once you get those hubs set up, you can take advantage of [indiscernible]. So I think there's a lot of things that we're doing around that, working with some of the customers that have really kind of come to the table after the flood that Steve talked about that's really paying good benefits efficiency wise. Rocky?

And I think that cross-functionally, Dave and I working together on the business process, I think on the sales side, the management of customers' week-to-week demand and polls, et cetera, we've made a lot of improvements to really get down and focused on those key customers that really move the needle. And working with manufacturing, I think we've really improved the business process of the supply chain with the customers and holding the customers accountable for their commitments. And I think it's resulted in that efficiency you've seen in inventory. It's a continued battle every week-to-week in the volatility to the market we're in. But I think that, again, we made some really fundamental advancements in our cross-functional management of the supply chain through customer fulfillment.

And we just have to be more disciplined. One of the things we're doing internally, Jayson, is we're actually changing our sales comp programs to be rewarded on basically, forecast accuracy and linearity, which is not something that typically we had done, not certainly at the level or priority that we're going to do it now. And the other again is just working with customers. I mean we had a customer last week that made a big change order on enterprise from one product to another. And we try to explain to them why that was difficult. And the response was, well, it's just a mix change. As if you just wave a wand over these drives and somehow make them something different. So it's just a different dialogue that has to occur that look, if you order one of these things, then we're building it for you and you're buying it. And if you're not, then there's going to be a penalty associated to it, not that we'll just swap these things out, like you can go to the hardware store and decide you want 1/4 inch screws instead of 3/8. So a lot of work the industry has to do, I think, to got the right discipline in.

Our final question comes from Cindy Shaw from DISCERN.

I've been hearing from some of my industry contacts that they're seeing the enterprise systems ship with just less capacity of drives and more empty drive space. And we can speculate as to a number of reasons as to why that might be up -- going on. But I was wondering if you can comment on, are you seeing that? And if you are, what do you think might be behind it?

I don't know that we're seeing that. But what I guess would be behind it is customers that hope that they can buy directly instead of buying racked-out systems that they're paying a huge markup on disk drives for. But I don't know that they can really do that, because if they're buying from OEMs, the OEMs really control that through their maintenance and service contracts. So you can't really buy a system and then buy drives from Seagate if you're buying a system from a traditional OEM because then your service contract typically is invalidated. So no one's willing to do that. So we haven't really seen that. But I do think that if you weren't talking about OEMs, I mean, I do think, architecturally, that's happening, that people are starting to design systems where they can incrementally add the storage they need. If those aren't really necessarily through the OEMs. I think it's -- that's more some of the new architectures. All right. Well, on behalf of the entire management team, I'd like to thank all of our employees and customers and partner suppliers and shareholders for your support and commitment, and best of luck to everyone on the East Coast as you recover from the storm. Thanks a lot.

Thank you, ladies and gentlemen. This concludes today's conference. Thank you for participating. You may now disconnect.