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

All right. Good. Welcome back, everybody. I'm Joe Moore. Very happy to have with us today the management team of Wolfspeed. We have Neill Reynolds, CFO; and Tyler Gronbach, who runs Investor Relations. And I'm Joe Moore. So maybe if we could -- I think most people probably know the company, but if you could just give us a little bit of an overview.

Sure.

We'll get into Q&A.

Yes. So we're a silicon carbide-based power semiconductor company. I think -- as Joe said, I think most people know. And I think we're right in the center of -- I guess, some people call it a revolution, but I think of it as kind of a renaissance of the power semiconductor market. And the reason for that is just the transition from internal combustion engines to electric vehicles, and that ramp is just happening a lot faster and a lot more heavier than I think that most people know. And silicon carbide plays a very important role in driving efficiencies into electric vehicles. So what we've seen there is just a really big step-up in demand and volume requirements, I'd say, over the next 10 years and had a pretty significant market growth rate over that period. And what that volume has done is it's not only increased electric vehicle rates or silicon carbide rates for electric vehicles, but also now, we've seen a big expansion in industrial and energy markets that's also kind of spawned that on. So as we look over the next 5 to 10 years, it's really become very much, I would say, in the industry, not just for us, a capacity game. How quickly can people bring on capacity to meet those top line type of targets? And from our perspective, the fastest way we're trying to do that and meet that demand is bringing on a new facility in Mohawk Valley, New York, where we do have a new automated 200-millimeter fab we're bringing online. And we're also building out our 200-millimeter supply chain. We anticipate that fab will be coming online soon. So I think it's really what it is about now and as we think about power semiconductors is meeting that demand over the next 5 or 10 years is going to be the big challenge.

And Joe, I think it's -- for those that don't know the company really well, we embarked on a journey about 4 years ago to drive the transition from silicon-based products to silicon carbide-based products. Why? Because it saves energy. It's more efficient. And we really see that in the EV space. But where -- as Neill rightfully mentioned, it's also expanding into industrial. So we really have this fit-for-purpose approach to the market. That's why we partnered with the device guys to sell them wafers. At the same time, we're also in the device business. So it's important to know that.

Yes. That's helpful. I wonder if you could talk a little bit just about the TAM in the markets that you're going after? I know you increased that number by a couple of billion in the last quarter, since the Investor Day. Where are you deriving those TAM numbers? And where do you see the biggest delta in the last few months?

I think that if you look at the market size, it's -- I think it's very -- from a power device perspective, it's actually changing very rapidly. We were having a discussion here earlier just about that. So it's really hard to get a really good handle on the size of the market. What we've kind of projected is to be north of $7 billion going up even above that as you get out by '25, '26 in terms of the power device market. However, I would say even in the last 6 months, that has stepped up even another degree, I'd say, step function more as we're seeing more and more transition on to electric vehicle platforms. And I think it's really driven by a couple of things. One is, I think there's -- it sounds -- it feels like there's a more rapid development on push to EVs and adoption rates than we've seen before. And the second thing is that silicon carbide within those -- within that acceleration is becoming more and more of a player. And I'd say for the deals that we do today is really about silicon carbide versus silicon carbide to get to the end of these deals. And the reason for that is most of the architectures are going to higher power, and higher power plays really well in terms of efficiency for silicon carbide. I don't know, Tyler, if you have any other views on that?

Yes, and I think that going back to the beginning and the journey, I mean, we've been driving down the cost of silicon carbide over the last several years. And what that has done is it's opened up the aperture for the market. So while we could spend the rest of the time here talking about EVs, it's important to also mention what's happening in industrials. You're seeing things like watercraft and vertical takeoff and landing vehicles. And refrigeration moving to silicon carbide, because we've reached a price point now where the efficiency is so strong to design that product in and move away from silicon, and that's really where this TAM, Joe, as you rightfully mentioned. It continues to grow, and it's well above $20 billion now, and we expect that only to continue to go up.

Great. And the biggest piece of this is going to be the inverter in a battery-powered car. It's actually a lot. It's close to half of the silicon content of a battery-powered car, I think, in some cases, -- that's a big number. Is it a fait accompli that this is the way of solving those problems and people have talked about silicon IGBTs and gallium nitride and those kinds of technologies, again on silicon? How do you see the trade-offs? And is it -- as I said, is it kind of at this point to you kind of done that this is the right technology?

Well, first of all, I'd say we're always paranoid about all technologies, whether our competition in whatever form it might come. But I do think that it is transitioning in a way that certainly at higher power applications. So the way silicon carbide works is the higher the power, actually, the blocking capability, the RDS(on) of that type of device actually performs better as the power structure goes up. So I think from that perspective, clearly, silicon carbide is becoming more of the, I would say, the product of choice for electric vehicle owners. I'll be leaving actually tonight. It was great to be here with you today, and then I'll head out on a red eye to go meet Gregg Lowe, our CEO, in Europe, and we're going to continue to meet with various OEMs and Tier 1s and customers around all of these different types of elements. And what I can say is that silicon carbide, as it relates to the efficiencies that we're seeing in the EV, certainly seems to be picking up traction. So like I said before, I think the transition to EVs is going faster, but I'd also say that the silicon carbide content in a vehicle. And by the way, Joe, I think that the -- if you look at an EV, about 80% of the content from a power perspective is based on the inverter. And then you have DC to DC conversion, and you have the onboard charging was a little lower percent. So the biggest chunk of the EV, which is silicon carbide-based is really where silicon carbide plays the best.

The total semi electric content of the car? That's a large portion.

Yes.

And then I guess where you guys -- I don't think I've ever covered a company that did both supplies and devices before, but obviously, there's unique characteristics in this market where a lot of the value is on the sort of material side. How do you view that trade-off? And just to the degree that some of your competitors have sort of said they all have their own internal sourcing programs because of this dynamic, how do you sort of see the trade-off start?

I'll get started. Maybe Tyler, you want to hit on it a little bit as well. First of all, -- this is -- it's an important point. And I think what you've seen over the last couple of years. I mean clearly, you mentioned it, silicon carbide is a notoriously difficult material to work with. We've been doing it over a long period of time. Actually built an LED business over more than 20 years, all on silicon carbide. So we've got a lot of experience in this. But I think one thing that's clearly changed, particularly over the last couple of years, is that most customers are dealing with a silicon shortage. And I think control over your supply chain from device all the way back through materials is very important. So if you think about a lot of these customers that are going through the silicon shortage right now, they're looking for supply capability all the way through from, like I said, device all the way back through materials. And I think that's what we're seeing here in this market. So I think part of it is you want to have that expertise in owning that entire supply chain, but a lot of it's about assurance of supply as it relates to customers and how they're viewing that transition that they are making, which is significant from internal combustion vehicles over to electric vehicles.

Yes. And I think, Joe, we've talked about this. But I think vertically -- to be vertically integrated, you're going to learn a lot going through that process. But I also think kind of where the market's going, we all want to kind of build our experience base. But at the end of the day, we have rivals, okay? We're all kind of competing in the same business. But I think if you ask Neill or I about how we think about things, anything that's a silicon carbide win is good. So even when we don't win, but somebody else does, and it's silicon carbide, that's kind of how we're thinking about the market today and for the next several years. Because guess what, capacity -- demand is going to outstrip supply. So as far as we sit here today, I think it's good for the industry and good for all of us, the more silicon carbide we can actually sell.

Yes, that makes sense. And then I guess, barriers to entry on the material side, to the extent that most of your customers do have some of their own plans, you've been open about that. But they've also come back and then extended for multiple years with you guys. So how difficult is this, and how long is your view -- your visibility into that?

Yes. Like you said, some of our rivals who we sell and devices that we sell materials to have made, they're going to go internalize and look to make their own capability. And I think that makes a lot of sense, first of all. I think if I were in their shoes, I'd probably do the same thing for a lot of the reasons that we just talked about. So as you pick that up and you start to look forward, I think the verticalization of that starts to make sense. I also think, though, that over time with the market picking up and the levels of increase in capacity that we're seeing, as Tyler kind of points out, I think we're just going to see that's a good thing for everyone. As I said, in the last 6 to 8 months, we've really seen another step function up in terms of the opportunity from a market standpoint. So I think from a capability standpoint, people will start to do that. Now it is very difficult to do. And they're talking about bringing on a lot more capacity in a relatively short amount of time. Silicon carbide needs to be growing at north of half the temperature of the sun. So it's a difficult crystal to grow. And then do it without any type of micro fractures or anything else, and it is also difficult. Remember, dealing with power chips that are pretty big. So defect density when it gets to yields also comes into play as well. So it's a difficult material to work with over time. It's the second hardest material in the world, so cutting it becomes a challenge as well. And again, we've had a lot of experience in doing this over time. So when you look at what we're doing, we'll probably get to this, but from a model standpoint, one, is we want to build a model that says, "Hey, for all of our kind of rivals in the device business, we understand what they're doing. We want them to be successful, and I think that's fine, and that's great." I think the second thing that we've got to be able to do then is to be able to scale our own capacity, and I think that will be good for the entire industry. But we do recognize -- and our numbers are built on all of them being successful, I think, which is great. But in the meantime, I think there's a long way to go in terms of bringing that capacity online and a difficulty in doing it to meet the demand that the market has. So I think it's not just a capacity equation as simple as throwing some money at it, I think there's a lot of learning cycles that are required in order to bring that capacity online because of the level of difficulty kind of working with that type of material.

Okay. Great. I want to shift a little bit to talk about some of the metrics that you guys give. You talk about the opportunity pipeline, which I think you've said is over $20 billion. And then the design-ins, which you had $2.9 billion last year, $1.6 billion last quarter, if I have the numbers right. How do you sort of define those numbers? And how much visibility do you have? Do we need to probability adjust some of that $1.6 billion that you won last quarter in terms of timing or things like that?

Yes, generally, yes. I've been in this industry quite a few years. Gregg has been in the industry, our CEO, for over 30 years. And as you know, Joe, it's a fairly common metric to use design-ins. Pipeline is something we used originally because before we had a lot of designs or design-ins and we started this a few years ago, we had to give out some metrics to say what is the size of the market? What is the level of activity? I think over time, what you'll see is we'll go from pipeline to design-in to design win pretty standard. But I think as you look at any embedded semiconductor company, there's always going to be some following of what happens in terms of your design-in capability and how that transitions to revenue. And what I can say is from a design-in perspective, the demand that's translating into revenue from those design-ins is outpacing what we've put out there in terms of revenue targets. So clearly, the demand is north of the $1.5 billion we put out there in 2024 and demand is higher than what we have in '26 as we follow those things through. So -- and I think that you could argue you could see that across the industry, and I don't think that's an issue. I don't know, Tyler, if you've got any other comments?

Yes. No. I think on the opportunity pipeline, that's -- think about it this way. It's current year plus 5. It's a combination of mid- and high-powered applications that might be silicon today, but we believe can flip to silicon carbide. And that's why it's -- when we first started out, it was about a $9 billion opportunity pipeline. Now it's north of 20. And that's only because, like I said, as that cost curve continues to come down for devices, it provides a bigger opportunity. So really, what we're quoting or what we're telling you in that number all the time is where we see a line of sight for silicon carbide devices to play.

Great. And you announced General Motors as a customer last year, and we haven't seen a lot of those yet in terms of the actual auto makers. You knew a lot of Tier 1 relationships. We haven't seen the actual automakers call that out. So I guess, a, what insight does that give you into the way some of those traditional automakers are thinking about hybrid versus battery power? And then b, do you think you could see more of those announcements down the road?

Potentially. I mean look, we work -- if you look at the design-in levels that we've had, we've had a lot of different agreements with Tier 1s, with OEMs. Not all of them were willing to put out a press release on different things, although we try to work with that [ it ] gives folks visibility as to what's going on. And I think if you look at General Motors, for instance, it was an insurance of supply program. And what we like about that, it really aligns both of our interests. They give us some visibility as to what their volumes are, and then we come back to them and we give them some assurances around how much supply we can give over time. And I can tell you, if you look at -- you mentioned EV versus hybrid, we don't really see a lot of hybrids. We did an update a couple of years ago, thinking from an emission standpoint, particularly in Europe, where were hybrids going to go, and they just couldn't meet the emission standards leveraging hybrid type of technology and the things that move to EV, and I think that's really what's happened. So I think more and more as these OEMs are going through this transition, and as I meet with them, it's a very significant internal shift for their company. The engine -- a nice engine -- internal combustion engine really was the technology picture of their company, right, with how well they could do that. So as they're moving over to e-technology and drivetrains. Now competitiveness in this arena is becoming more and more important. So I think as you look at this over time, competitively as well as from an assurance of supply standpoint, silicon carbide for them really makes those convergence of those 2 things very, very important.

And Joe, I think the other thing to add on there is, although we haven't had a chance to talk a lot about our wins publicly, some of our partners like BorgWarner and ZF have done a really nice job in the market and have talked about where their inverters are getting placed into vehicles, and they're both a partner of ours. So it's another way to kind of think about how we might be winning in the market.

Okay. Great. So maybe turning a little bit to your own progress. Maybe start with Durham. You obviously, going back a year ago, struggled a little bit with gross margins out of that facility. You're seeing improvement. Where do you stand now on sort of generating the -- I know the big improvement comes from Mohawk Valley ramps. But where do you stand operationally with the existing facility?

We're making progress. I mean one of the things that you look at our footprint today is that we've got this small kind of undersized fab in Durham. It was an LED factory that we sold the LED business and as part of that began to outsource a lot of that volume and then invest over 100 tools into the factory to bring up the essentially silicon carbide MOSFET type capability over time. And given the structure of that, doing kind of automotive capable MOSFET in an LED factory, was more challenging than we thought. We brought in new management, and we are seeing some improvement there. But as you mentioned, Joe, I think we'll see some improvement over time. I think that takes us through this kind of timeframe of transitioning to Mohawk Valley. I want to call it a stop gap, but it's -- clearly, it's keeping us going until we hit kind of a bigger ramp rate with a bigger diameter and a large modern fab, we should continue to see some improvement. Like I said on the call, I think over time -- over the next -- the shorter period, what we'll see is revenue growth, particularly driven by the power device business, that's at kind of a little bit more of a lower margin mix just because it's running out of that maybe lower cost footprint, 150-millimeter type capacity, that will offset some of the improvements. So we'll kind of see maybe kind of flattish type of margin, maybe modest improvement until we hit Mohawk Valley. And then when we hit there at 200-millimeter modern fab, you start to see the flip over in terms of the margin capability.

And maybe talk about that a little bit. I mean you'll have the first 200-millimeter fab. It's a big footprint for sort of the size of this business today. And how quickly do you see the benefits of that phasing in?

So I think the way to think about that is we should be running wafers there soon. We had a pilot line up in SUNY Albany in New York. And that pilot line was demonstrating 200-millimeter capability over time. So we built out the supply chain and ready to over the last couple of years for 200-millimeter. Since then, that pilot line now has been dismantled and it's been -- being installed into the Mohawk Valley clean room kind of as we speak, may actually be finished with it. So what we'll do now is we'll go run wafers. They'll take some wafers from the line. We'll put them from the pilot line that we've kind of pre-produced, put them into the Mohawk Valley line at different stages and then start that line, and we'll start getting learning cycles off the fab. We'll then take those wafers in those learning cycles, and we'll start giving those to customers for them to do their qualification, integrate the systems, and we'll work with them on that. We anticipate that kind of qualification cycle with customers picking up steam over the rest of this calendar year. And then as you get into the first half of next year, we'll start to see that actually translate into more meaningful revenue. So we could have like revenue before that, but I think from a meaningful perspective, kind of moving the dial. I think of the back half -- first half of calendar '23, this time next year as being kind of that transition point where we start to see Mohawk Valley make a difference. And I think, in that sense, the more revenue that comes out of Mohawk Valley, not only will the revenue come up, but I think we'll start to see the margin improvements as well.

And your revenue confidence, I guess, is -- goes hand-in-hand with the confidence of bringing up that much capacity. You've talked about $1.5 billion in 2024, $2.1 billion in 2026, with growth in both the material and the device side. What drives the confidence to sort of forecast out that far and invest on that point?

Yes. I think, Tyler, maybe you want to jump in here. I think that the design and trajectory and how that's translating to revenue, clearly, that's driven upward pressure to our 2024 and 2026 number. So this really is not a demand situation where we're concerned about top line. In fact, I think from an industry perspective, this is really about capacity and how quickly this type of technology can ramp. I think the days of questioning, do we have enough revenue, is there enough design-in activity? How does that translate to revenue? It's really a bit of a moot point right now based on what we're seeing. So it's really about how quickly can us and others bring on capacity to support a market that's growing this rapidly.

Yes. And I think, Joe, if you go back a couple of years, when we first started talking about design-ins, we were talking about $0.5 billion worth of design-ins per quarter. In our most recent quarter, it was $1.6 billion. So I think -- what that tells you is, a, we've got traction, which we feel really good about, which leads us to why we feel good about the $1.5 billion for 2024. And on that 2026 number, we know there's upward pressure. And I think we've mentioned this a little bit earlier, but capacity is going to be the key, right? Neill and I have been talking a lot about that today, but how fast can you bring up capacity? How fast, just to remind everybody, we've only built out Mohawk Valley to 50%. So there's another half of the clean space that we can fit out there. And that's really how we think about things and the opportunity as it relates to upward pressure in 2026.

If I could just add to that quickly, Joe. I think as you -- we've had 30-plus years of experience. We've done ramps in silicon carbide and LED businesses. And our perspective is that -- and how our team talks about it, who has a lot of that experience, is you want to be really methodical, at least we do, about how to bring that capacity online. You're talking about a notoriously difficult substrate to grow and cut. And it will be at this kind of level of capacity for the first time ever. And on top of that, doing it in a fab that requires [ runs ]. So really, in our mind, you want to get learning rates from the substrate all the way through the device, get the learning rates, understand it, ensure we're shipping good quality material to our customers. Also underpin the margin capability we talked about. And as we methodically get that capability, we'll expand the capacity to meet the demand. So again, a lot of confidence on the demand side. I think we just got to be very methodical in bringing a technology to market in a way that we're confident in terms of what the results are going to be.

Great. So all of that, I guess it comes to the gross margin targets of getting to the 50% by that 2024 and the kind of 50% to 54% longer term. Again, can you talk about your confidence on that, assuming the revenue numbers arise?

I feel great about it actually because I think that we're here sitting in the kind of mid-30s today, but we do have -- we haven't shipped anything on 200-millimeter and typically in silicon semiconductors and you make that diameter change from 150 to 200, there's a couple of things that happen. One is you get 70% more good die per wafer. And that generally transition -- translates to about a 40% cost reduction. However, we also believe we're going to get, and we're seeing that in the pilot line, and we had seen it in the pilot line in SUNY Albany I mentioned earlier, 20 to 30-point better yield performance. And on top of that, we'll put it in a facility where the cycle times will be dramatically better than what we have in Durham today. So we do have the advantage of having -- even though I think over time -- in the early periods that silicon, a 200-millimeter substrate will be at a bit of a higher cost. We'll get a lot of benefit in the fab from cycle time yield in that wafer diameter change. So the more revenue that we push out, the better it will be. Now part of it is we're working off a base in Durham that isn't that great right now. We kind of say one way to have really great numbers in Mohawk Valley comparatively not to have good ones in Durham, and we're improving that. But when we see that transition. So if you look at perspective, if you look at just kind of benchmarks of what we need to go do, we don't need to go do -- build a factory in Mohawk Valley that's better than everyone else or something like that, we just have to do it about on par with what everyone else is doing and do it in a state-of-the-art automated factory. So we feel pretty good about we know how all that comes together.

Great. Let me take questions from the audience. I have several more, but if you have any questions from the audience first. Anyone? I do have one upfront.

[Technical Difficulty]

Yes. I'll repeat the question. The question was just 200-millimeter supplies, are you planning on being in the market?

Yes. So I get -- I get this question a lot. It's interesting because let me just take a step back for 1 second. If you think about the size of the market that we're driving and clearly, over time, we've seen this in silicon. People don't want to transition from 150 to 200 millimeter, and we get that. But we have a unique situation going on. First of all, if you look at the cost per millimeter square of a 200-millimeter substrate versus 150, in the early stages of ramping that when we have less volume, it's going to be more costly. Now the benefit of a 200-millimeter wafer generally happens in the fab. You run at the same amount of time over a similar cost [ of rent ] and you get a lot of those benefits. More die per wafer plus you get the benefits of driving through the fab. So our benefit on the margin side is really by taking it internally into our fab. However, if the cost of that substrate is still relatively high, it's not necessarily commercially viable just yet. So let me talk about, we'll make that decision in a couple of years. Well, it's literally -- once the 200-millimeter wafer is commercially viable and the cost is at a place what would make sense for a customer to run it in their fab and get those benefits, and we would consider that. But right now, we're just -- I think our focus is let's drive it through our Mohawk Valley facility. That volume will help us bring that cost down, and then that decision will come kind of come in time.

But if your materials customers wanted to build a pilot and figure out how to do 200-millimeter and they can't buy it from you. I mean that does sort of force their hand to try to develop it more on their own?

I think it's a possibility, Joe. But I think our view is that we just want to have the flexibility to kind of go manage that, and we'll look at it over time. And our focus really needs to be around developing it for the Mohawk Valley and filling that today. That and of itself is a pretty big hurdle we've got to get though, we got to bring this new fab up on a new technology. So we're really focused on that. So while that -- I agree with you, that could be the case. I think we have to stay focused because what you're talking about now is accelerating a business plan on top of what we have on our plate today, which I think would be an added challenge. So I think right now the best thing to do is to ramp a 200-millimeter, get the volume benefits from that and then put it into our Mohawk Valley fab, and that decision could come in time. Remember, a lot of those customers are ramping on 150-millimeter today. That will be a pretty quick turnaround to go to 200 that quickly. I think there'll be some time and space to go figure some of those things out.

From a volume perspective, they need a fab, right? They need to build [ a 200 ] fab but just -- I'm thinking more... pilot line ...

A lot of them are on -- going around 150 and 200 fabs, probably, which I think is fine. But when we talk to them, I mean, there certainly is interest. But think about it as well. I mean if you're selling to automotive customers, you've got to do multiple qualifications. One of the benefits to us are running 200 originally out of Mohawk Valley as those customers have to be one qualification cycle. So now we don't have to really worry about that. If you're doing 150 you're ramping now and then you got to switch it over 200 later, you got to do another qualification cycle and you're driving more cost into your customers. So that's another kind of disadvantage of doing it that way. So I think some space probably makes sense. But I think in the meantime, people are going to debate this and talk about it. But I think, look, take the step back and look at the overall market, I think getting volume and capacity into the system is the most important thing right now.

Great. Any other questions from the audience? Yes, they do have one more here.

Just a quick one. If you saw demand running substantially above where it is now, how quickly could you turn on the rest of that 50% capacity in Mohawk Valley? I assume it's not the same as the original build?

No, it's funny you say that because if we were trying to build a fab today and do it under these circumstances with COVID and supply chain challenges it would be a lot longer, but it won't be that long. So typically you think about -- I think about a fab expansion you want to do about -- you want to pull the trigger kind of like a year in advance or so even when you have the clean room built. Clearly, you've got to watch lead times on equipment and things like that, and we're doing that closely. But again, I'll go back to it. I think it's really for us about bringing up that factory in the supply chain from materials all the way through devices and ensuring that, that's in good shape. This is a challenging -- like I said, it's a challenging thing to do. Silicon carbide is a new technology in that sense and ramping at these levels, I think, is going to be challenging. So you don't want to be -- we don't want to be flippant in terms of addressing that, okay? I think that's kind of like step one. Secondly, we went out and did a convertible offering. We've got about $1.3 billion in cash. You can [ look at ] the math after the convertible. We think we'll shift to operating cash flow positive next year. So the profitability situation will be better. We've been really investing in R&D. We'll do $100 million, $200 million in R&D work this year on 200-millimeter. So -- and if that happens, we would -- and things go well, we would try to accelerate that. So we have a lot of dry powder, I think, from a capital perspective on top of that from the state of New York. We've only received about $60 million of the $500 million of reimbursements. So there's another $440 million to come our way. So if you think about the amount of cash we have in the balance sheet, the reimbursements we're going to get from New York, we can go affect that. And now it's going to be a matter of methodically managing this, looking at lead times and balancing those 2 things to fill it out. But if we have the opportunity to address that, we will. And if we did, we would come out and talk about how we could go do that and what it would mean. So in the meantime, we've got to run a wafer in the fab and sell to someone. And as soon as we do that, we'll come back and talk about what the inside potential to be.

Joe, what I might add is just the leading indicator of the thing to listen to or 4 on our future calls is design-in wins. That will kind of dictate how fast we might build out the second half of Mohawk Valley.

Great. Well it brings up the end of our time. Thank you guys very much.

Thank you, everyone.

Thank you.

Thank you.