Western Digital Corporation (WDC) Earnings Call Transcript & Summary
June 1, 2021
Earnings Call Speaker Segments
Karl Ackerman
analystHello, everyone. Welcome to the Cowen 2021 TMT Conference. I'm Karl Ackerman and I am a senior analyst covering semiconductors with an emphasis on the memory, RF and power semis categories. I'm pleased to present to you the team at Western Digital, including Bob Eulau, CFO; and Peter Andrew, VP of IR. Gentlemen, thanks for being here.
Robert Eulau
executiveThanks, Karl. I appreciate the opportunity.
Karl Ackerman
analystThank you. So the presentation today will be a fireside chat format. If you'd like to ask any questions during the presentation, you may submit questions on the portal or you may send questions to me directly as karl.ackerman@cowen.com. With that, I'm going to turn it over to Peter to read a safe harbor statement, and we'll go from there.
T. Peter Andrew
executiveOkay. Thank you very much, Karl. Today, we will be making forward-looking statements, and I ask that you refer to our SEC filings for the risks and uncertainties with these statements. We will also be making references to non-GAAP financials, and a reconciliation of our GAAP and non-GAAP results can be found on our website. So back to you, Karl.
Karl Ackerman
analystGreat. Thanks for that. So to start, we've been asking all of our companies at the conference to discuss the level of visibility each has into the second half based on order backlog, inventory on hand and continuity of supply, which has become a larger challenge for many, I think, in general. So I was hoping you could tell us regarding your ability and strategy to meet demand over the next few quarters, particularly given recent restrictions by the government of Malaysia, where I think you do have some facilities there. So if you could just kind of highlight your visibility in the back half that would be helpful.
Robert Eulau
executiveOkay. I think lots of questions in there, Karl. So let me talk about Malaysia first, so I don't forget about it. Malaysia, we're at full production. We've been dealing with COVID, as you know, for well over a year, we're viewed as an essential business there and we're operating at 100%. And you never know for sure, but so far, things have gone reasonably well there. We're also working with our supply base to make sure that we're not impacted in Malaysia. So far, knock on wood, we're doing okay there. As we look at the overall demand picture, we've been excited about 2021 for quite a while, really dating back to the October time frame. We saw transactional markets behaving well. And by that, I mean our retail market and our commercial distribution channels, we were seeing prices go up. Hence back then, we haven't seen it yet in the OEM market, now we're seeing it in the OEM markets. So the year is playing out essentially as we had expected. As you know, on the Flash side, there are lots of opportunities for continued growth with 5G out, with our new products in the enterprise SSD space. And now the fact we really have a broad portfolio where we're strong in client SSDs as well as in retail. I think we've got a very good offering, and it allows us to move bits to where we think we can get the most margin at any particular point in time. So we feel really good about Flash as we go through this year. We think demand will grow somewhere in the mid-30s and we think supply will grow a little less than that and should set up for a favorable second half of the year. On the hard drive side, again, a fairly similar story. We knew we were going through a digestion cycle. Second half of last year, we ended up with our revenue down significantly in the September and December quarters. And then we recovered as expected as we entered this year. We've got a great set of new products out with our 16- and 18- and 20-terabyte family. And this past quarter, we think we bottomed in the hard drive market. We had fairly even shipments in exabytes across 14, 16 and 18. We're expecting in the June quarter that we will be shipping primarily the 18-terabyte drive, and that will be the exabyte leader for us. So we think it's shaping up for a good second half.
Karl Ackerman
analystThat's very helpful. Appreciate that, Bob. I guess you alluded to it in your last remark in terms of pricing. But I know NAND pricing is something that we certainly cannot forecast. But in general, there have been some rising input costs from the -- from a substrate level, from the logistics side of the house. And so I was wondering if you could just highlight your ability to at least offset those rising costs or those elevated costs today, both across your hard drive and HDD business. I think if you could just kind of highlight how would you characterize the pricing environment today that would be helpful.
Robert Eulau
executiveYes, I'm definitely happy to talk about that. I think the other thing you alluded to in the prior question was the supply chain challenges and we definitely are navigating through those right now. We've seen tightness in terms of the controller chips that we use, both on the Flash side and on the hard drive side. We've seen, as you noted, other input cost pressures coming in. And I think most of us believe we're in an inflationary environment right now. So the key for us is to make sure that, first of all, we can get supply so we keep shipping. I think we've done a nice job there. And then I think over time, clearly, our gross margins are not in a sustainable place. So we need to be able to get margins up over time. And I think on the hard drive side, we'll begin to do that. We have -- you may have a question about it later, but we have had some short-term demand on the GSI, which has impacted primarily our channel and retail businesses, but we've been able to move price up there some, which is helpful. And we'll see what that means for the overall business as we move forward. But I think on the hard drive side, it's probably going to behave more like the Flash business as we go forward. It's kind of in an equilibrium in terms of supply and demand. There's not a lot of incremental capacity coming online. So we'll see how that plays out in terms of pricing. And I know from our standpoint, to make incremental investments, we really do need to see gross margins improving on the hard drive business. We still also, I should note, have the headwind due to the COVID-19 costs and the lack of passenger commercial flights, carrying freight coming out of Asia has really caused our freight cost to be up quite a bit. And it's been about 1.5 headwind each quarter for about the last year. And we don't see that changing anytime soon. On the Flash side, from a cost standpoint, we're really pleased with how we've been doing. Our long-term goal is to have 15% year-over-year cost declines. And we've been -- we overachieved that in the last couple of quarters. We may have a quarter where we don't get to 15%. But I think long term, 15% is a good number for us. We feel really good about how we've done with BiCS4, which is our 96-layer product. We're just now beginning to ramp our BiCS5 product, which is our 112-layer product. And some people say, "Well, you have 112 versus 128, that's actually a good thing." What it means is that there's -- we need fewer tools. We have faster cycle time, less capital intensity and therefore, we should have lower costs. And we think it's likely will have higher yields as well. So we feel really good about BiCS5, which we're just starting to ramp now. We expect as we get to the end of the year that we'll have more bits on BiCS5 than on BiCS4. So that transition is just starting, and that will help us sustain cost declines as we go out over the next year or 2. So -- and then I guess the other thing on the Flash piece is we're also impacted by the controller chips on the Flash side. Now we design our own controllers and we go to foundries and they're in -- capacity is in tight supply there. But we've been able to get most of what we want. And then to the extent we're short on controllers, we've really been allocating them to higher capacity points where we can get more revenue and more gross profit per controller. So hopefully that got to your question.
Karl Ackerman
analystYes. It did and more, so I appreciate that. Maybe just moving to end demand then. I guess, what is your sense of end demand today from hyperscale customers, both for -- if you could talk on both for nearline hard drives and then also enterprise SSDs? Because on one hand, a lot of companies across the supply chain have said, "Hey, things appear to be picking up" I guess. But on the others, some would say, "Well, is there a pull forward of demand from these programs that you've actually announced and expect to ramp over the next few months," right? You've got a couple of cross hard drive and a couple of across enterprise SSD. So if you could just talk about what you're seeing there, that would be helpful.
Robert Eulau
executiveYes. I mean, frankly, we're seeing strong demand from the hyperscalers on both the hard drive side and on the enterprise SSD side. I think probably there's more demand there than we can actually fulfill right now. So I don't know that there's a lot of double ordering going on. I think it's a pretty well-orchestrated supply chain there, and we'll do our best to meet the demand as we can going forward. But it's also what we expected, as I said. I mean, we knew we were going through digestion at the end of last year. We knew this year was going to be strong. And we still believe the second half is going to be stronger than the first half. And we've got the supply we need to deliver on the guidance that we gave and fulfill most of the needs of our customers.
Karl Ackerman
analystAny -- no, I appreciate that. Any thoughts on, I guess, small, medium business in enterprise accounts? I ask because we've been hearing about a server cycle refresh. And this inventory digestion cycle in the second half of 2020 has been setting up some easier compares. But any thoughts on -- your thoughts on non-hyperscale cloud demand over the next couple of quarters or what you're seeing there?
Robert Eulau
executiveYes, we've been monitoring that pretty closely. And I would say we believe that the -- we call it the on-prem market has really stabilized at this stage. I'm not really seeing incremental demand at this point. Hopefully, a server refresh would help be a catalyst for that. We also believe that employees going back into the workplace has the potential to be a catalyst for that. So I think now that we've kind of bottomed out on the on-prem business, the enterprise business, we should be setting the stage for some growth as we move forward.
Karl Ackerman
analystNo, that's helpful. I guess that kind of leads me to my next question, which is, would you say signing LTAs is a relatively new industry phenomenon? I ask because one of the investor debates arguing margins for hard drives will remain suppressed. Is that the pivot to enterprise? Just concentrates the number of customers procuring drives and that the penalty for not taking one of a handful of $1 million dollar hard drive orders is extremely high versus kind of the price and getting volume. But I'm wondering if signing LTAs allows you to embed maybe a price component? Or maybe more importantly, lessens the desire by industry participants to engage in some irrational price behavior that's happened in the past. So anything that you could highlight in terms of that dynamic changing would be really helpful.
Robert Eulau
executiveYes. I think the hard drive industry is definitely at a critical stage right now. I think it's in transition. The reality is we probably shouldn't have been pricing product every quarter with the major customers. And so I think we're getting out of that mindset. The long-term agreements we've done have been typically 2 or 3 quarters. And what we're really trying to do at this time is really stop the decline in terms of pricing. And it's our intent, particularly with the 18-terabyte drive to really maintain a price that's fair and that we deliver some TCO benefit to our customers. At the same time, we see our margins recover because the margins are really not at an acceptable level from a 16-terabyte drive perspective. And that really makes us think that the industry can invest in capacity. So I think margins need to go up. We're, as I said, pretty balanced in terms of supply and demand right now in the hard drive industry, and to invest more, I think margins need to be better. So we'll continue to work on that. So I think the LTAs are good. I think it is changing the slope of the price decline curve, and we'll see as things get tighter what happens with price.
Karl Ackerman
analystGot it. Okay. No, that's helpful. Maybe pivoting to NAND for a moment. Your emphasis on introducing NVMe-capable enterprise SSDs over the last 2 years, I think it certainly put you in a much more competitive position and allowed you to regain some of the share loss that you had in 2016. So I guess, could you talk about the progress you've made on broadening your qualification pipeline? And maybe has -- I guess, has Intel's sale of the enterprise SSD business allowed you to compete for more bake-off opportunities, maybe that you otherwise wouldn't have had?
Robert Eulau
executiveYes. I think one of the real synergies that we have in our company that we've talked about is the go-to-market. And so with all of these hyperscale companies, we already have significant relationships on the hard drive side, and it gives us the opportunity to compete first slots on the enterprise SSD side. So we had success earlier this year in terms of getting involved with a couple of the hyperscalers, continuing to work on it and others. Also selling our NVMe drives into the channel, which is helpful. So I think, overall, that business is going quite well. And what it really does is round out our portfolio. So now we have 4 areas of the market in which we have over 10% of our bits going. So we have the enterprise SSD space. We have a client SSD space, where a lot of times we've been #1 or #2. We have the mobile space, which generally has been a little under-indexed, too. But it's still over 10% of our bits. And then finally, we have the retail business, which in most countries around the world, we're #1 in retail. So we've got a very balanced portfolio. And what it allows our team to do as we move forward is to really prioritize where we place those bits, now that we have good solutions in each of the markets. And I think we're already starting to see the team do a good job with that. So the enterprise SSD business was critical to build out that portfolio. Now we've got that done, and it's a matter of just making sure we prioritize the right markets.
Karl Ackerman
analystYes. No, that's helpful, Bob. I guess that you had referenced in your opening kind of comments in terms of supply and demand in NAND. If I wanted to go back to that, I think some investors are concerned or it would be easy to point to the fact that, hey, maybe NAND profitability may moderate in the second half as some of these Korean peers have been a little bit more vocal about outgrowing bit demand for this year. And I guess, what sort of initiatives have you taken or can you take to respond to such a scenario?
Robert Eulau
executiveYes. So I guess, first of all, what we try to do is really model what actually happens rather than what people say is happening or may happen. And as you know, our stated goal is to grow with the market to maintain the market share we have. And we think over an extended period, we will achieve that goal. You never know when you hear from competitors, I mean, what is their baseline. Maybe they went through a period where they didn't have as many bits. Now they're ramping up and they have more bits. We're going through nodal transitions at different points in time. There -- we have different -- as I mentioned earlier, different companies have different capital intensity in terms of a given node even, we're usually on the low end of that. So you have those. And then yields, you never know what yields are like at your competitors as well. So there are multiple reasons that can influence the deployment of CapEx. And again, we really try to monitor in aggregate what the market is doing and then make sure we're growing our bits in line with the market.
Karl Ackerman
analystGot it. No, that's helpful. Again, I guess, then focusing on cost declines. Your NAND cost declines have been very healthy, I think at roughly 20% year-over-year during the March quarter. You've made some pretty strong progress on BiCS to date. Your long-term cost improvement rate is 15%. But I guess, why wouldn't your cost structure remain ahead of your long-term model when we reach -- I think you mentioned you're going to reach big crossover of BiCS5 later on this year. You're going to be launching BiCS6, which I think you had a presentation a month or so ago that talked about 70% more bits per wafer. So I think if you just kind of highlight your NAND cost declines, both near term and longer term, that would be very helpful.
Robert Eulau
executiveYes. I mean, we have to strike the right balance on each of the nodes. So we're not going to all of a sudden move everything from BiCS4 to BiCS5, and that would flood the market with bits, it wouldn't be good for anyone. We also have to have the controllers in place to take advantage of the new nodes. So we're working through that and making sure that's synchronized. So I think our plan at 15% is a good, solid, achievable plan. And you're right, I think we've overachieved in the last couple of quarters. But there may be a quarter or 2 where we're under 15% year-over-year as well. So again, over the long term, I think 15% is a good number, and we'll be able to grow bits with the market and bring costs down along -- around that 15% number.
Karl Ackerman
analystOkay. That's very helpful. And just maybe last one on NAND, and I'd like to pivot back to the hard drive business. But your manufacturing partner, Kioxia, has not yet raised proceeds from an IPO. Some investors have asked, "Hey, does that actually limit Western Digital's ability to fund NAND CapEx and progress toward future node migrations?" And so if you could just kind of highlight your ability to fund necessary CapEx regardless of what your partner does or does not do? Could you discuss that, please?
Robert Eulau
executiveYes. Well, we do invest together. And so we want them to be successful, whether it's going IPO or whatever they choose to do. They have a very strong ownership group. They have a lot of access to capital. So I don't think it's going to be a dilemma. I think that we're united in terms of our goal to grow with the market. And I think that's what you'll see out of the JV. So I just don't think it's an issue. It would be crazy to let that become an issue.
Karl Ackerman
analystYes. Okay. No, very helpful. Maybe just moving back to hard drives for a moment. I guess, how do you feel about your hard drive margins in the back half of the year? And maybe if I could go out on a limb, into 2022, assuming you probably got a pretty good mix of 18-terabyte. And maybe if you could also just kind of highlight why your cost structure on 18-terabyte is better than 16-terabyte because I think that is a little bit lost by some investors?
Robert Eulau
executiveYes. So a couple of questions in there. And first of all, we're very bullish on the 18-terabyte drive. As I said, I think it will be a very strong exabyte shipper this quarter. And I think it will be the flagship product of the industry for the next few quarters. So I think we're in a really good place with that product. Then and as I think I mentioned before, I mean on the 16-terabyte drive, they're more of a price taker, we weren't able to set the price. We think we've been able to set an appropriate price on the 18-terabyte drive, so we can deliver a good total cost of ownership to our customer. At the same time, we can improve our margins for our company. And I think we're in a good place there. The reality is the cost structures are not that different between the 18- and the 16-terabyte drives. So when you have a tough price on 16, it makes margins a challenge. So I think over the next few quarters, we should see a sequential improvement in gross margin. That's our goal. Eventually, we want to get back to 30% as the first achievement. And then over time, we really think the capacity enterprise business for the industry needs to be in the mid-30s in terms of gross margin to invest in the kind of capacity that's necessary. We went through many years where we were downsizing on the client side, moving assets and resources from the client hard drive business to the capacity enterprise business. And we're to the point now where that's not going to be happening as we move forward. So we'll still see some decline on the client business. But the increase in the capacity enterprise business is going to more than offset that. And we really think the hard drive business is going to be a growth industry as we move forward. So it's an important inflection point for the industry.
Karl Ackerman
analystYes, it is. That's -- I appreciate that. I guess to follow up to that, the quick question is, well, how -- what drives us toward improving margins? And then I guess, maybe the second follow-up to that is my actual question, which is that as we discussed your technology road map, we haven't heard much from you lately in terms of your energy-assisted technology road map. You used MAMR in the past. I guess, if you could highlight what's the largest impediment to ramping MAMR today? And it sounds to -- some investors would say, "Well, maybe we improve margins as we grow beyond this capacity point function of 2 terabytes." And so what's the -- how do we do that? And then secondarily, what are some other initiatives you can do to continue to save costs as you use both CMR and SMR technologies?
Robert Eulau
executiveYes. So we're -- the 16-, 18- and 20-terabyte drives are energy-assisted hard drives. And energy assist is a technology that we developed as part of our investment in MAMR. And we continue to invest in MAMR and HAMR, and we'll use innovations out of those investments in the products that we bring to market over the next few cycles. And we think we've got -- definitely have more life in terms of what we can do with the energy-assist product line before we have to do things like transition to glass, et cetera, which add more cost to the product. So we'll continue to have a very cost-effective product. And then over time, we'll continue to bring costs down. Right now, on the 18s and 16s, we still have opportunity to improve our yields. Yields are on a typical curve. But as we improve the yields, that will help our margins as well. And so we feel very confident with our ability to bring costs down. And then we can -- we have to slow down the rate of price decline, which is part of why we did the long-term agreements that we talked about. And I think that will help us gain in terms of gross margins as we move forward. I really think the last quarter will prove to be the bottom. And as I said before, you'll see sequential improvement as we move forward from here.
Karl Ackerman
analystGreat. Now that'd be welcome and would be great to see. Before we let you go, Bob, I guess, what do you think are the 1 or 2 things that investors may have underappreciated about the Western Digital story that you'd like to highlight before we go?
Robert Eulau
executiveYes. I mean, I think we've just gone through a major product refresh. So we're very strong in terms of starting to ramp on BiCS5 on the Flash side that I mentioned before. And we feel like we've now got the full portfolio of products that we can take to market and we can capitalize on the opportunities in multiple markets as we go forward. We also, as we were just talking about, have a new product lineup on the hard drive side. So it's a really good set of products that we have at this point. And then obviously, the economy, at least, in the U.S. is certainly picking up. And hopefully, globally, we'll see progress against COVID and see economies coming back, and we think it sets up for really good year. I'd say the other more tactical thing that I think people underappreciate is the value of the joint venture. It's a joint venture between Kioxia and ourselves. It's been around for over 20 years. The only reason a JV like that lasts is because it adds value to both parties. It gives us a great economies of scale, both in terms of R&D as well as in terms of production. And so I think we really benefit from that at the cost structure level, which we're talking a little bit about earlier. So I think overall, the market conditions are good. We've got a great new set of products, and I think we've got a structure that makes us very, very effective. We think we get a lot of synergy out of our go-to-market. And we didn't talk much about that, but we've benefited over time by selling into PC OEMs and OEMs in general. We've been able to transition from client hard drives to clients SSDs, more of a substitute there. On the data center side, we think enterprise SSDs and capacity enterprise hard drives are complementary, and we'll be able to grow in both of those areas as we move forward. So I think there's -- that wasn't a short answer, but I think there are some things there that perhaps people don't pay enough attention to.
Karl Ackerman
analystGot it. That's really helpful, Bob and Peter, I really appreciate the time today, and thanks for joining.
Robert Eulau
executiveAll right, Karl, thanks a lot. We appreciate your time.
Karl Ackerman
analystThank you.
Read the full transcript via the API
You're viewing the first half of this call. Get the complete Western Digital Corporation transcript — plus 251,000+ transcripts from 12,000+ companies, speaker segments, AI summaries and full-text search — through the EarningsCalls.dev API.
Get the API View API docs →This call discussed
For developers and AI pipelines
Programmatic access to Western Digital Corporation earnings transcripts and 251,000+ others is available through the
EarningsCalls.dev REST API. Plans from $24.99/month — full transcripts, speaker segments,
full-text search, and the recently-added /api/v1/transcripts/recent polling endpoint for ETL pipelines.