Wolfspeed, Inc. (WOLF) Earnings Call Transcript & Summary

Great. Welcome back, everyone. I'm Joe Moore from Morgan Stanley semiconductor team. Very happy to have Neill Reynolds and Tyler Gronbach, CFO and IR, respectively, for Wolfspeed. So just quick safe harbor. For important disclosures, please see the Morgan Stanley research disclosure website at morganstanley.com/researchdisclosures. If you have any questions, please reach out to your Morgan Stanley sales representative. So welcome, guys. I think most people probably know you up to a point, but maybe just a couple of minute intro into the silicon carbide market and Wolfspeed enablement of that market.

Yes. So the silicon carbide market, I mean, what we're seeing right now is a pretty dramatic change, historic change in high-power semiconductors. And it's really driven by the advent of electric vehicles, the adoption of electric vehicles, where silicon carbide plays a really important role in terms of that efficiency. This is something we've been doing for a long time, over 30 years, but now it's really finding its home and driving efficiency in electric vehicles. And what that's doing is it's providing a volume basis and capability here over time that will drive a lot of efficiency, a lot of volume, eventually bring costs down and really proliferate that technology across the power semiconductor landscape, including areas then outside of electric vehicles as well, which actually make up about 30% of our sales pipeline on top of electric vehicles. So it's a huge opportunity here. We play in both the materials of the crystal growth of actually growing silicon carbide crystals. But we're very much focused on the bleeding-edge technology at 200-millimeter for the device as well. But essentially, the role that we play, we're kind of in the heartbeat of that drivetrain and electric vehicle, and that's really opening up a lot of different applications.

Great. Yes. So maybe we'll talk a little bit about the market size and then talk a little bit about your role, sort of a dual supplier of supplies and devices. In terms of the market, it's very clear that this is an accelerating trend that everyone is moving to get more EVs to market, more battery-powered vehicles to market. It's also still kind of a gold rush feeling from a device standpoint, where there's so much activity and you have -- we can add just the top 5 device manufacturers, including yourselves, we get to $3 billion this year, which is kind of a heady number. Can you talk to how you get comfortable within that kind of gold rush type of market that you're shipping to in demand? You guys have talked about this being potentially sold out for another 6 or 7 years. How do you think about that?

Well, I think it's one of these cases, Joe, where -- and Tyler, maybe you can comment on the market size as well, where all boats kind of rise with the tide. I think what we're seeing is these adoption rate in electric vehicles is happening so quickly. And we're starting to see this gestate over time in terms of what these ramps are going to look like that it creates a tremendous opportunity. So when you look at the sales pipeline that we've generated, I think we started out at $9 billion or something. We started measuring this 3 or 4 years ago. It's now well over $40 billion. And that's not just on what we compete on, that's what we compete with other people on. And in that same time frame, we've done over $16 billion of design-ins. And as you said, right now, we are supply-constrained, both in the short term and even as we look out into the long term, and we see that across the industry. And if you look over time, the market's going to grow at roughly a 40% CAGR. We see that potentially out to 2027, and even out to the end of the decade, where today, it's a much smaller market. We could see it being $18 billion or $20 billion over -- as we get to the end of the decade at a higher adoption rate of vehicles.

Yes. And I think to build on what Neill just said, we're at this inflection point. And Joe, you know this because you've been covering us for a little bit now. We've talked about capability and scale, and we're on the verge of kind of turning that on. So we said on our last call, we're at an inflection point as we bring on the world's first 200-millimeter device fab. And then last September, we announced we're going to build the world's largest materials factory in Siler City, North Carolina. So for where we sit, it's what Neill just said, it's this opportunity pipeline that's rapidly growing. At the same time, we're trying to scale capacity to meet market demand, but it's still very early stage.

I mean it's kind of unique to me, and I haven't covered you guys for that long, but I've covered automotive semiconductors for a long time in various forms. And it will take some time, several years to qualify a $2 analog part. And now you're in this position where every automaker is rushing to market on the highest-valued component of the car relatively quickly. And I think there's a certain amount of uncertainty about the whole supply chain. And I think -- so have you guys -- does it seem like, to you, that the pace is as frenetic as maybe it seems to us? And do you think there's a risk of -- that we're double ordering along the way as people look at, hey, I'm uncertain about these supply chains?

Yes. I don't see the -- I mean, there's -- obviously, there's always the potential that people are double ordering because you're supply-constrained, okay? That's pretty normal in semiconductors. But what I don't see, though, is if you look over time -- if you look at the automotive customers, for most companies, this is somewhat of a life or death situation. They're taking the R&D budgets for internal combustion engines, and they're really targeting that towards electric vehicles. So there really isn't something else they can go to. They're going to take their platforms, they're going to change them over and move them over to electric vehicles and not really have something else to sell for the most part as you get out in time. And that's what I think is really driving at the end of the day the freneticism, as you call it, Joe, is because they really need to know, they need to be successful on a number of vectors in a very new supply chain that's coming online. I remember sitting here last year with you, and I think the takeaway was this industry needs capable supply over time. I don't think that's changed at all. Investment and supply is what's going to be needed because we're dealing with the largest automotive customers in the world. We're going to want to transition from their internal combustion engine vehicles over to electric vehicles where silicon carbide plays an important role. Now silicon carbide is a difficult thing to bring online. We're -- and the way that we're attacking that is by going to the -- as you know, to -- from 150-millimeter wafers to 200-millimeter wafers so we can bring more supply. That's 70% more surface area for everything we deliver at a 40% lower cost. So it really solves 2 market issues in a very short amount of time. By bringing more supply to the market, but also driving some cost efficiencies over time. I just start to think about bringing on more and more volume. But yes, I think if you look at the OEMs, the Tier 1s, they're all managing through this transition. I think capable supply, steady supply is something that's going to be needed. And that's obviously why we're investing heavily in this area to bring on that type of capability, and we feel that's going to be needed across the industry.

And Joe, if I could just add on. I think going through the pandemic and the stock-out that we saw on silicon products, I think, is what's driving a little bit of this pace of play by OEMs and Tier 1s to where assurance of supply is really one of the primary drivers of conversations when we talk to customers. And so you probably -- many of the people here in the room know about the deal that we did with BorgWarner, which provided a capacity corridor of up to $650 million annually in devices. We've struck a partnership with ZF. And there have been other things along the way. But I would say that we're coming out of this period where automotive manufacturers weren't able to make cars. And I think now what we say it is as they look at silicon carbide is a critical element of the drivetrain. They just want to be able to have assurance that they're going to be able to make the cars that they plan to do.

Great. So I wonder if you could talk a little bit to the content per vehicle that you guys see, and I'll follow up. I know Tesla raised some questions about this last week, I'll give you a chance to respond to that. But can you talk -- before we get to that, just what do you kind of see as the general content for the traction inverter and other silicon carbide?

Yes. So our TAM model, as I said earlier, gets us to about $18 billion to $20 billion of market size for silicon carbide devices out in that kind of 2030 time frame. That's at about a 40% EV adoption rate. And what we look at is about $300 to $500, $600 per vehicle in terms of silicon carbide content opportunity that's in there as you look over time. That's basically what we use to kind of come up with what market size is. Now when you get -- when you dig into that a little further, I think the interesting point is that when you break down that -- the market per vehicle, 85% to 90% of that is in the drivetrain. So it's in the module and in an inverter, and that's really important because that's really where silicon carbide plays the biggest role and the more high-power part [ of the end ]. So we're really focused on the area that's -- that high -- the highest part of the content also being the area where silicon carbide drives most of the efficiency.

And so then coming to the Tesla comment from last week, they talked about a reduction of 75%. I think they were talking about on future vehicles, not every vehicle. But maybe you could just talk to that. And I don't know if you want to go too specific in any one customer, but do you see that content changing over time as people become more budget-conscious and bring these into lower-priced vehicles?

Well, clearly, people are very focused on the drivetrain, and they made some comments last week. And there's a lot of speculation around that. People are speculating that we could be talking about a very kind of low-cost, a low-power vehicle at maybe a 400 voltage -- 400-volt architecture. And what that means for us is that it probably strengthens our TAM, and I'll explain why in a second. And the reason for that is when you talk about the $18 billion to $20 billion, I mentioned, if you look out in time, at a 40% EV adoption rate, if you double-click on that and look at what that -- bifurcation of that market is, we assume that about 60% to 65% of that market is silicon carbide EVs. And the rest, the other 35% to 40%, would be silicon-based. So what that means is if you're talking about the very low-end power part of the market, which you could say -- is silicon carbide going to be used at really low power because it's known for really great efficiencies of higher power. We're probably expanding the TAM for the silicon carbide penetration part of that. So it probably strengthens the TAM over time. And when we work with our customers, we are generally seeing, from our point of view, is really a transition to a more -- either multiple power but an 800-volt architecture and the reason we can only use silicon carbide. And the reason for that is you get the benefits of silicon carbide at the higher voltage levels. So regardless of where people go, if this is the way that this starts to head or whatever it might be, regardless of what that is, it's probably good for silicon carbide because there's going to be more of it, which then brings us back to, we probably need more supply, industry needs more supply, and we need to really support what this transition to electric vehicles is going to look like.

Okay. Great. And then you talked about the content in the car. Can you talk about the content sort of beyond the car when you think about charging infrastructure and things like that? And also your guy's capability to serve that market with -- in areas like GaN and other technologies?

Yes. So let me get started. Maybe, Tyler, you've got a couple of comments, but I'll start off with this. What's really interesting is silicon carbide has been around for quite a long time. We've kind of pioneered this technology over many, many years. But it had to find its home from a high-power perspective when volumes for electric vehicles came online. And what that's done, though, now is that the higher volumes is it's created an opportunity to create a lot of volume. And that volume will obviously take down cost as you start to expand this capability. So as you look at this over time, that's now opened up a number of opportunities for different platforms and different applications that you can use outside of electric vehicles, which have been terrific. And maybe, Tyler, do you want to talk to what some of those might be?

Yes. I mean as Joe points out, charging infrastructure in a vehicle, it's $300 to $600 worth of content. In a supercharger that charges an electric vehicle, it's about $1,000 of silicon carbide content. So charging infrastructure is going to be an important part of our strategy as we go forward. But as Neill points out, what's happened, automotive has driven the cost down of devices. And what that's done is it -- what it's done is open the aperture for industrial and energy for things like vertical takeoff and landing vehicles, train and traction, e-mobilities. And we're seeing things like -- concepts like green taxiing, where they want to take existing aircraft fleet and convert it into a hybrid structure for battery-operated taxiing to and from the gate. So we see as cost come -- as the cost comes down on devices, the opportunity across a wide spectrum of sectors is really an opportunity for us, too.

Great. In terms of customer announcements on the device side, you guys have spoken to General Motors and Jaguar Land Rover. On the OEM side, you've talked about Tier 1 relationships with BorgWarner and ZF. I guess it's hard for us to analyze from the outside what that market share looks like when you talk about it by customer. How do you think about those wins? How exclusive are they? Do you -- are there customer -- obviously, there are customers that aren't announcing wins, but like how much of this is you pushing people to making announcements versus you people -- you not necessarily wanting those announcements? Just how do you guys think about that?

Well, I'll let Tyler talk to what -- he deals with this almost every day. But clearly, we're working with a number of large OEMs, Tier 1s, as you mentioned. We've been working with General Motors for several years. We've announced that. Great partnership with Jaguar Land Rover. And most recently, Mercedes, who we've talked about as well, not to mention some of the Tier 1s. So we worked with these various customers on announcements when it makes sense for them. But at the end of the day, we're really a supplier to them working on trying to solve solutions. And where it makes sense, we'll obviously engage with them in terms of what those partnerships look like.

Yes. And I think we're a newer entrant into the market. And so -- and we're just a pure silicon carbide play. So for us, we are opportunistic and try to announce whenever we can. But we also know that there's some element of -- customers are not comfortable maybe making a pronouncement with us. And we're okay with that because -- I would say this, we've had over $16 billion of design-ins over the last 4 years. You should think about 60% to 70% of that is in the automotive space. And what that probably tells you, we're probably punching above our weight a little bit in terms of automotive. So we feel -- even though we'd love to be able to disclose more, we're very happy with kind of how we're trending in automotive.

And in terms of those customer wins, how do you think about the capabilities that you guys have? I think from our perspective, you're so clearly the strongest provider of supplies of wafers that the vertical integration is a very powerful message. But if we talk to your competitors, one of whom is going to be down the hall in 20 minutes, they'll talk about the capabilities they have within modules, the fact that they're already in the silicon IGBT market. They've done all the automotive certifications and things like that. And the customers seem to care about both of those things. So how do you -- how would you assess your capabilities at this time? You mentioned sort of seeing yourselves as a new entrant. How are you going about building out those capabilities?

Yes, I think customers look at a lot of things when they look at a new opportunity. And I think some of the things you mentioned are valid, by the way, Joe. And certainly, we look at our rivals in the sector. There's very good companies out there who have been working in automotive for quite a long time. But when you look at us, you're looking at a technology that's ramping for the first time for something that's very, very difficult to do, starting with the substrates. And I think that we have very, very deep domain kind of expertise in those areas. When you look at the device side, so when you look at the -- I don't think there's any real question as it relates to us regarding the materials capability in the substrate and crystal growth capability. So then the question comes back to devices. We've been a fully integrated device manufacturer for more than 10 years, all the way going from crystal growth, all the way through a module. We've been doing this for a long time. In fact, we came out with the first silicon carbide MOSFET back in 2012. I think maybe one other company, I think it was ROHM, and maybe right around that same time period. So felt like we have deep domain expertise in the crystal growth in the material side, but I would also argue very deep and capable domain expertise in the device side as well. And I think our devices are highly capable and always score very, very highly from a performance perspective when they're tested by customers. So we feel very good about our technology capability from crystal growth to MOSFET chips and even through modules where we've been also engaged over more than 10 years. So we've had this kind of -- and I think we were the first fully integrated supplier in the industry all the way from crystal growth through the module. So I think when customers look at us now, the question is not so much around technology capability, I think we always score pretty high there, the question is scaling, bringing this to market at scale. And what we've done there is a couple of different things. One is we brought in a lot of people who have capability around automotive semiconductors who are working with us. So our Head of Quality, Lisa Fritz, she's got many, many years of experience in automotive quality at another company. So when they see her at the table, they feel very confident in it. The second thing is that we're bringing on more capacity than anyone in the industry. We are -- we've just gone through a silicon supply shortage for a long period of time, and we're going to grab the largest silicon carbide manufacturing footprint in the world, and we're going to do that at the bleeding-edge technology at 200-millimeter. And when a lot of people talk about being able to deliver and being able to divest, we're actually doing it. They can go to Mohawk Valley. They can see a new fab. They can see we're starting wafers at 200-millimeter and bringing that to market and bringing on capacity in a big, big way to be able to support them as we get out into the second half of this decade. So I think those are the pros and cons. And I think when they look at us having deep domain expertise and really a significant amount of capacity at the bleeding-edge technology, that's what they're really looking at when they're looking at us.

Hey, Joe, one more point on this. I think the way that we look at the industry, we're still so early stage in silicon carbide. So we know that some of you might say competitors, we look at them as rivals, right? We want to continue to see everyone win in the marketplace with silicon carbide. So as Neill pointed out, we're a supplier to our rivals on the wafer side, and then we're competing for a device business. And so as we sit here today, we kind of know that there's not enough capacity coming online, that's why we've talked about the investment of over $6.5 billion over the next 4 years to build out capability. But for those of you that have looked at our Investor Day, that only gets us to about $5 billion of revenue on a TAM that we think is about $18 million to $20 million. So I think, Joe, there is a lot of business to go around, but there's also a lot of capacity that still needs to come online.

Absolutely.

So you mentioned Mohawk Valley. I think maybe you could talk about that a little. You're going from a sub-margin facility in Durham that was sort of built for LEDs, migrating to the world's only 200-millimeter silicon carbide facility. How is that going? Can you just give us an update on how you're ramping that fab? And can you talk about how that impacts your economics over time?

Sure. First of all, the transition from 150-millimeter kind of legacy technology over to 200-millimeter is a big shift. This is something we've been working on for a number of years, and it's something that we're -- the team is really, really excited about, to be at this point. And where we're at is we're bringing up a couple of things right now. One is we're bringing the Mohawk Valley fab online. The data coming out of the fab for the testing we've done looks very, very good. At the same time, we're also bringing up 200-millimeter crystal growth in volume. Now we've tested this over a number of years. We've built many, many 200-millimeter wafers. We've run them through our pilot line that we had running up in Upstate New York for a number of years. So we have really good confidence in the technology that we're bringing up. The next step for us is to now start to bring this to fruition, start to bring this volume on -- bring this capacity on at volume. And that's really the next step of what we're working on right now. So the way I tell people is, we're -- the Durham fab you mentioned from a device perspective, that's about $400 million of revenue per year. It's probably capped right now at about $100 million a quarter. So what we need to do now is transition over to Mohawk Valley. We should see maybe single-digit millions of revenue as we get to the June quarter here, and that all looks good. And then it's going to be a question between now and the end of the calendar year, that first kind of tranche of millions of dollars of capacity for a $2 billion facility. Now there could be a little bit of bumpiness as you start to bring all of this together at the same time. And I do anticipate that could happen between now and the rest of the calendar year. But I think as we start to bring the utilization up in the fab, we get more cycles of learning and we get more cycles of practice at bringing up this capacity, we'll see that come on in a big way. And that's -- people ask about revenue growth and margin. I think I've said on the last call, all roads lead to Mohawk Valley, and I think that's really the question. I think we'll start to see a ramp-up from Mohawk Valley. We'll get the cost benefits of the 200-millimeter substrate. We'll also see a big spike up in revenue as we start to see it come online. But we'll see how it plays out here as we start to bring up the initial [ lots ] of utilization.

And can you talk about gross margins at maturity in that facility?

Yes. So as you look over time, obviously, our target model's between 50% and 54%. We'll obviously be using that for maybe the Mohawk Valley facility for our most advanced technology with our biggest customers over time. So I would anticipate that there's a big differential in not just cost but yields and cycle times and whatnot as you get into the facility. So you could see the margins coming out north of 50%, for sure, in line with our target model over time as you start to think about that. I think that's worthy of maybe stepping back and really thinking about that for a second. Mohawk Valley cost us about $2 billion to build. We got $500 million back in our deal with the state of New York, it's about $1.5 billion net. This is a facility that will generate $2 billion annually in revenue with about -- with north of 50% gross margin and about a 60% cash fall-through. So the economics on it are really terrific. And I think it really plays a role for us in a number of ways. One, it creates a really high-quality, capable factory to serve automotive customers. But from an investor perspective, it creates this really wonderful cash generation vehicle that will really serve us for many, many years.

And the 200-millimeter aspect of it, do you -- could you characterize the margin improvement just from that? Because some people have talked about -- some of your competitors [ that sell ] millimeter wafers can be quite a bit more expensive because they have to ticker and things like that. Like obviously, you're doing it because you see a cost advantage in doing so.

There's no question. We can just talk about basic semiconductor math. 150-millimeter wafer going to 200, that is good, okay. So let's just make that simple statement right there. Now over time, it will take volume to get cost out of the substrate. That's just natural. The first substrates will cost more than the ones when we have a lot more volume. But I do anticipate over time we'll get the cost per centimeter squared on an equivalent basis over time. The unique advantage we have is that we'll be putting it into a fab. The way that you get benefit from going from 150-millimeter to 200-millimeter is driving the benefit through the fab. You get 70% more surface area for every math level that you run through the fab. And that's the benefit that we're going to see regardless of the substrate costs being a little bit higher early on in the plan. So that's all baked into our assumptions that early on, it will be a little bit more costly. But of course, we'll get the benefit of running that internally, now paying a market for it and then getting the benefit as you run that -- run those substrates through the fab.

Okay. Great. So I'll ask one more question and then I'll open it up for the audience. You talked about addition of 2 new facilities in North Carolina and Germany. Can you just talk through what that means for your numbers over time and sort of what it means for your scale down the road?

Yes. It's -- we talked about it earlier. We're building the largest silicon carbide manufacturing footprint in the world, and we're going to do that at the new kind of 200-millimeter bleeding-edge technology. And we're going to do that in a couple of ways. Tyler talked about a $5 billion fab device footprint. That's what we're building. So the Mohawk Valley fab to do a little over $2 billion, $2 billion or so of revenue at maturity. The fab that we announced recently in Saarland can do -- is about 25% bigger than that. So think about $2.5 billion, maybe a bit north of that. And we still have the Durham fab doing $400 million. So there's a lot of capacity to bring on over time to get to roughly a $5 billion device opportunity by 2030. Now in addition to that and just as importantly is the substrate capacity that needs to come on to support that. Mohawk Valley, when we think about it, the Durham facility can generate substrates, maybe probably around half or less than half of that fab, so additional capacity is needed. So you have to double that manufacturing footprint for crystal growth and substrates over that period. Well, if you have another fab after that, you need to be able to double it again, except that fab is bigger, so it's more than doubled again. So we're building out The JP, named after our late Co-Founder, John Palmour, in Siler City, North Carolina, that will have the ability to scale to serve these multiple fabs and even fabs beyond that. So really, it's really important for us to have all of these pieces coming together to achieve a full kind of integrated supply chain for silicon carbide to get out into the second half of the decade.

Great. Let me pause there and see if we have questions from the audience.

Yes. Two questions, [ the one would be on the device ] side. The design of the device, so planar versus tranche, can you elaborate a little bit where you stand on how important that is in order to win business? And the second one, I mean, you're building a lot of capacities. Others are building a lot of capacities. You're talking about $18 billion TAM. What is the risk in the industry to overbuild in terms of capacities?

On the first one, on planar versus tranche the capacity a little bit, on the -- just on the -- when you think about silicon carbide, it's following a similar path from silicon in that perspective, where everyone has a road map. And what customers really care about is what I talked about before, it's -- they look at a myriad of different things. They look at the device cost and price, they look at the performance, then look at your ability to supply it. So whether it's a certain type of design isn't so much important as it is what is the performance and the price that goes into that particular application as well as your ability to serve that supply. So I think it's really a mosaic of things that come together as it relates to what customers think about as you look out over time. From a capacity perspective, and maybe, Tyler, you've got some comments on this as well, what we see right now is a supply-demand disconnect. And we anticipate seeing this out to the end of the decade. So our view isn't so much -- and my concern when I wake up in the morning isn't about is there oversupply in the market, it's quite the opposite. Is there enough supply in the market? If you look at the amount of capacity that we're putting in place over time, we really like to see that across the rest of the industry to see more capacity come on. Tyler talked about some of our -- some of the other folks being more rivals than competitors. And I think it's very early in the stage of growing this industry to really be talking about that because it's going to take all of us to bring this capability online and supply it over time.

Yes. And I think where we sit today and you look at the practical application outside of automotive, that is definitely going to drive a lot of demand. But we had an OEM executive tell us a short while ago, we're going to claw our way to 30% penetration on EVs, right? But then at 30%, the cost of the internal combustion engine car goes up because of the fixed cost on things like mufflers and spark plugs. So his cautionary tale to us was at 30%, the -- it flips and be careful because penetration rates on EVs are going to run a lot quicker than what it took us to get to 30%. So as we sit today and kind of look at some of the industry estimates, and we also have great line of sight because we work with some incredible Tier 1 partners that have insights to all the models coming to market, this is where I think Neill and I come back to, there's -- the supply and demand mismatch is going to go for the next several years.

Maybe we could just wrap up with on the supplies side of the business. You mentioned over the time that all of your customers have plans to bring up some of their own substrate capacity. On the other hand, you've had the renewal at STMicro. You had a recent renewal, which we think is ROHM. I don't know if you said that.

It was ROHM. Yes, we did confirm that was one.

So people are still renewing, people are still coming to you for supply. So how do you think the strategy of being vertically integrated and selling supplies to people that are your rivals, can you talk about that?

Yes. So I think that -- this is important. We talked about being supply-demand disconnected for a long period of time, probably for -- out to the end of the decade. So -- and there's 2 things to consider here. One is the 150-millimeter kind of business that we're selling wafers on. So think about it as kind of the legacy 150-millimeter technology. And we know a lot of people are going to try and in-source that. We're a major supplier in the industry, and that's okay. We need the capacity in the industry. But if someone -- if one of our -- one of the companies that we work with make a road map that says they need to do X percent internal, they need to buy X percent from us, we'll work with them on a long-term agreement to help supply that. And we'll go and try and support the industry in terms of bringing this technology to fruition. But in the meantime then, we'll spend most of our time really driving the 200-millimeter bleeding-edge capability and continuing to focus on that, build our device capability and build that out, that capability over time. So I don't think of it so much as a threat, Joe. I think it's more about bringing this technology and advancing these to market and advancing it in such a way where it creates stable supply for the industry.

Great. Neill, Tyler, thank you so much for your time.

Thank you.

Thank you.