ON Semiconductor Corporation (ON) Earnings Call Transcript & Summary

Good morning. Welcome back to another session. I'm Vivek Arya from the BofA, semiconductor -- Semicap Equipment team. Really delighted and honored to have the team from On Semi, Hassane El-Khoury, President and CEO and Thad Trent, CFO, joining us this morning. I thought we would go through some Q&A. But if you have any questions, please feel free to raise your hand, and I can be assure to get you in.

So maybe, Hassane, I'm sure we've discussed the macro and so forth. But I feel that On has been on such an interesting and unique kind of transformation journey over the last right 2.5 years since you joined. So maybe it will be helpful for the audience to listen to where do you think you are on the transformation right aspect and how do you look at the company over the next 3 to 5 years?

Sure. So if you look at kind of the phases of transformation in order for us to be able to extract and deliver value, not just shareholder value but customer value as well, we had to focus the first few years or the first phase on structural transformation because we needed to establish the company as leader in the field and really have the operational efficiency in order to execute whatever comes next. So if you think about what we've done over few last few years, we rationalized the portfolio. We've done a lot of investments in key technologies like silicon carbides and silicon power. We shaped up the transformation of our image sensing, group as well, invested in silicon power, not just the high voltage. So from a portfolio -- and we exited some areas that we don't see the value that we can bring to some customers. So from that perspective, that's why we rationalized the portfolio. Right after that, we rationalized our manufacturing footprint, while we talked about fab lighter, where it -- having to really focus on scale manufacturing that -- where we can extract value, and we've divested for fabs as part of that. So that puts us in a very growth market with good portfolio, good technology foundation, and a very predictable and sustainable financial model. And that's what we've achieved over the last couple of years. Looking forward, we just had our Analyst Day a few weeks ago, is building on top of this strong foundation to deliver the growth, to deliver the financial performance, the profitability and therefore, shareholder value while we maintain our leadership in the technology to deliver value to our customers through leadership in silicon carbide, leadership and image sensing and a large $14 billion TAM expansion on a subset of the analog world, which is on the driver and controller that fit very well with our strategic intent on the power. So that's kind of where is left, while we do the fab right, a fab lighter to fab right, which is now take our existing very efficient footprint and make it more efficient by running better lanes or better manufacturing strategies to maximize the output and therefore, reduce the cost.

Got it. So we'll come back to the structural aspect, maybe just to get the near term out of the way. How do you see -- what's kind of your state of the union right now and in terms of looking at the demand environment now versus what your assumptions were at the start of the year? What's better or worse than what you thought before?

Yes. Look, I think I'll let Thad comment on kind of how it compared. But I think we are, call it, a very stable. It's better than it was, I think, in the fourth quarter if we look at pushouts and cancellations, but we're taking a very cautious approach. The term we use is were puckered up. We're cautiously optimistic. You see us running the company from inventory -- the things that we control, we have very good controls. As far as inventory in the channel, overall inventory, you see us taking down our utilization so we stay ahead of what the market does. And look, if the market goes sideways for the next few quarters, we're in a great position. If the market is better than we think in the second quarter, everything we've done turns into tailwind. Utilization will be a tailwind for margins. Inventory in the channel will be a tailwind for demand. So we are in the best place we can from the areas we control to navigate whatever the macro throws at us.

Yes. I would say and just to add on that, if you look at coming into the year, right, early last year, mid last year, you got consumer and compute gets soft, we saw consumer facing industrial get soft in Q3, Q4 time frame. And I think as we were looking at schedules at that time, they peaked in Q4. And so far this year, they've declined, which is a good sign. So I think as we entered the year, we were kind of bracing for what was going to happen in this downturn softness, whatever it is. I think sitting here today with reschedules actually declining. I think we feel like it's pretty stable, and we've got great visibility for the rest of the year. If you look at our LTSAs, we're still signing LTSAs up, right? And we just announced the TESCO last week. So the momentum from the customers in terms of engaging long term is still very positive. Whatever the short term is going to do, I think, as Hassane said, we positioned the company very well to come out of this, and we're continuing to manage puckered up. Utilizations are at 71%. We've proactively taken that down. I think we'll run the rest of this year in that range until we really start to see a recovery. But I think as we look through the rest of year, it feels pretty stable.

Got it. How much of that is ready -- Sorry, please.

[indiscernible]

Yes. So with China EVs, first of all, we're market share gainer, right? So we were late coming into the party there. So as we think about China EVs, it's all upside to us. We have seen softness. But if you think about it, we have -- we have a small base that we're actually growing on. So anything is incremental to us. But we have seen some softness. I wouldn't say it's extremely bad in terms of softness, but it is softer than what they were projecting at the beginning of the year.

Yes. It doesn't impact our trajectory as far as the growth, cumulative growth because it's going to go up quarter-on-quarter because they're still building. EVs are penetrating, but we're coming off a smaller base. Now some of our peers if you have majority share in BMS or anything like that, you can -- you will see fluctuations. We're not at that pace. So we feel pretty comfortable about our outlook specific to that.

Of course. Now specific to the automotive market, one thing I think you also highlighted at Analyst Day was the ability to have that diversity, right, whether it's on the silicon carbide side or IGBT side, right, or with image sensor. So maybe talk about that because I feel that it seems like people are making the whole auto EV market to just be, do you have silicon carbide or do you not? And it's going to be commoditized over time. So how are you looking at your engagements and how can you sustain the market share you have in silicon carbide?

Yes. So I'm going to cover it in two areas because specifically on the outlook of, oh, it's going to be commoditized. So let me cover on our positioning. The breadth of the portfolio is actually competitive advantage. We've said that 12 months ago or 18 months ago when we started the transformation that when we engage with a customer, we engage on a solution. Because we have silicon power and silicon carbide, we don't need to be in a position to go in and say, well, you have to use silicon carbide. And you're not shoving silicon carbide an opportunity that doesn't need it because then it looks like it's very expensive. So there has to a technology fit at a price point in order to solve at a system level price point. It's not optimizing on silicon carbide, which means if we're able to put silicon carbide, which is a more expensive material, but the customer gets a benefit on batteries and cooling and so on, the customer wins. So do we. That's a win-win. And in some areas, we were able to achieve that with IGBT or silicon carbide or IGBT and silicon carbide. And we are in a very unique position to be able to do both. Now that's on the power. And if you recall from Analyst Day, there's 500 more dots on that car. So once the anchor point, which is the inverter is decided then, okay there's the onboard charger. There's under-the-hood products that we do because when you go to EV, all of these loads that typically are run by belts in a normal engine, all these become motor controllers and discrete fats. We benefit from all of that. So that's the content that we talk about when we say we're going from an ICE car to EV car. And now as OEMs proliferate their platforms to more lower price vehicle, but much more volume, IGBT starts to play a big role. And we are, again, in the best place because we cover both, and we've been supplying both. So from an OEM perspective, they see us as the supplier of choice because we service all of their platforms. They don't have to go, well, if I have silicon carbide, I got to go from this supplier. Now I need IGBT I got to find another one, one-stop shop. And that fits. On the comment of commodity, we do not see anywhere close to this being a commodity. There's a lot of chatter from certain peers where the substrates may be a commodity, right, because it's going to be so broadly available. We're not in the market of substrates. I don't sell substrates in market. On Semi is a product solution player. So if substrates in the next 10 years get really cheap, guess what? If they get cheaper than what we can do them internally, it's still into our margin. Do you see what I mean? So it's not a detractor for the substrates to be available. As it stands today, I don't see it. We'll cover that because I mean we know the challenges that are to scale substrate business. It's one thing to deliver substrate. It's the other to scale it the way we have. And some of my peers will tell you how difficult it is. So I'm not worried about the commodity word when it comes to the substrate because if we say our silicon carbide is accretive to our margin with our vertical integration and somebody can do it cheaper than we can internally, great. I'll just source from outside we're even better from our margin. Do you see what I mean? So that's fine for us, and we'll keep targeting the market and we'll keep investing. But it's very -- also from a strategy perspective, you can't go one way or the other. We have internal. It is the strategic thing we provide to our customers. They value vertical integration. Because in what world am I or any of our sales team will go to a customer and say, yes, we got a supply resiliency plan. It's 100% sourced in China and offset.

Right. Got it. So on the having internal capacity, I think you've had that exiting this year, the goal is to be 50% internal or more than 50%?

Probably more than 50%.

More than 50% internal. Where does the move to 200-millimeter come in? I think you've also mentioned that some time, but do you think a competitor who has access to 200-millimeter has a cost advantage over On Semi?

I don't think so because to go to 8-inch or to do 200-millimeter, you have to do it right. And all I can say is who is better and who is not just look at the margin profile. Look at the margin profile and revenue and wins. It tells you who is doing it better. So talking about, well, 8-inch or 200-millimeter is going to be much better competitive advantage versus 6. Today, our 6-inch is yielding better than anybody out there. And our results show that, right, versus their results without naming names. We have said our 8-inch. We're on track to introduce our 8-inch in '24, so next year, with a production or revenue in '25. We've said that 12 months ago. We're still on that track and we're delivering to that. We're sampling. We're running 8-inch already today and our conversion to -- 6 to 8 is pretty straightforward because the fab we run 6-inch SiC in it is the same fab that runs 8-inch power today. So we don't have the challenge of ramping a new fab from 6 to 8.

So big CapEx?

No. It's conversion CapEx. And the conversion CapEx is primarily on the furnace side, where it's not even the furnace, it's just the heater core. Everything else is already 8-inch capable. That's why for us, when we're ready, we're just going to start running concurrently 8 and 6 and then over time, just keep converting to 8. That is the best from a risk mitigation, that's the best approach to ramp a conversion to 8 while ramping hard on the revenue ramp that we've talked about 70% year-on-year or -- 70% for the next 5-year CAGR.

Got it. Now one thing on the long-term supply agreement, Hassane. Autos is a dynamic market, right? EVs is even a more dynamic market. So if most of your customers don't know how many EVs they will produce in a given year, right, whether they will be successful or not, how can they give that certainty to you in terms of a LTSA?

Yes, because a lot of our LTSAs are platform wins. So maybe the customer doesn't know the blue model is not going to work, but the Model B is going to be better or the Model C is going to be introduced. If they're all on the same platform, the customer is able to change the mix, which is, again, why we, from a platform perspective, the scalability of our inverter design and our approach that Simon talked about at Analyst Day is actually competitive range because exactly that thing. Because look, the customer can introduce a vehicle and all of a sudden, a lot of people want to buy it at the expense of the next vehicle. They want to be able to sprint capacity and change the mix and our scalable approach to the design is actually a benefit for them to be able to do that because the worst thing you can do is, you introduce a vehicle, it takes off and you're constrained. You just killed your own success. So they want to be able to do that mix across, and we're able to provide that to them. That's a competitive advantage. That's why we've been getting the LTSAs. And that's why customers don't hesitate giving us the LTSA because they see that flexibility that they can use to their advantage.

Do you think that -- so we see a lot of OEM announcements, right? And most of them have tended to dual source, right? Right, you and your European right here in most situations. Do you think those are exclusive by platform? Or like are you exclusive on a platform basis or their dual sourcing on a platform basis?

It's by platform. So because you don't -- meaning it is exclusive by platform. Some of the wins, for example -- look, I'll acknowledge 3 years ago, we weren't in the market, On Semi was not in the market. So therefore, some of the designs that are ramping with certain OEMs are designed in 3, 4 years ago. So although you may see the same logo being announced by us or prior by one of our peers. Some of it is new platforms that the OEMs are introducing and some of them are platform shifts that we have seen. And we've seen that accelerate a little bit in the last 2 quarters. Just because OEMs are starting to realize, it's not a slam dunk, and they want to go with a slam-dung supplier that is able to supply. So that has shifted a little bit, too. Obviously, it doesn't change '23, because we've been sold out. But that -- from an LTSA perspective, that has given us a different outlook, a more favorable outlook. But it is because the customer is not going to say, okay, I'm going to swap modules on the same inverter because then you end up with a different performance, right? Because we have -- and I've always said we have better performance on a platform than our peers. That's why we win the platform. So having a dual source with somebody you just won against on performance, then you're going to end up with the same platform with two performances. So they separate by platform.

Got it. One last one on silicon carbide and competition. So one of your competitors, right, they're facing some challenges in scaling the material side, but they have this really nice large device factory. You think -- let's assume they are able to overcome those challenges. Then should investors be concerned that there could be a flood of capacity on the device side that comes into the industry? Or do you think that that's not a concern?

Look, that's not a concern where I sit because all of our designs are LTSAs, and most of the LTSAs were signed when they didn't have an issue. You see what I mean. A lot of LTSAs that we had beginning of '22 before all the issues from our peer started surfacing were irrelevant. You know what I mean? So we got the LTSAs because of the performance. Again, it's not because of supply, you have to meet performance. It doesn't matter if you can make all the silicon carbide in the world. If you don't hit the performance targets in order for silicon carbide to be justified versus silicon. Then it doesn't matter how many wafers you got. So we've been winning on performance. We've been winning on performance of our dye and the performance of our package together. That's a competitive advantage. That's why we've been winning. And that's why we have the LTSAs. They'll work -- I don't know what specific issues are over there, whether or not they work it out, that's their execution, but it's not going to change the trajectory we're in. I said it may modify a little bit on where customers go. Yes, we're just going to shift share. And that goes into our LTSAs, but it's not going to change whether and how fast and how aggressive we win. We put a target of 35% to 40% market share. That assumes not somebody failing. That assumes us delivering and executing, which I have full confidence we will execute.

Got it. And then on that market share goal, right, so I think you are sort of in that share right now, right, your European peer is in around that share right now. Do you think looking out 5, 6, 7 years that this becomes that kind of 2, 3 players controlling 80%, 90%? Or you think it's going to be a lot more fragmented industry?

It's not going to be fragmented. It's going to be a handful that navigated in the short term, call it, next 5, 10 years. EVs are only going to be 50% penetrated in total SAAR. So the runway, you can think about it as a 20-year runway by the time you get full EV penetration. So as that maturity and the penetration starts coming in, there's going to less and less. Look, there's one already starting to drop out, where they doing it all because look, the barrier of entry is high, unless you can scale and scale very quickly. You're just not going to -- you're just going to be a laggard, and in that business with the CapEx intensity, you can't be a laggard. The way we're doing it, we got best-in-class ROIC. We believe in the way we have the model to invest. We have always said, we invested in brownfield, which is more efficient. Our ROIC shows it. Our gross margin shows it. We're performing financially better than anybody else in that market. And all that matters and the sustainability of that investment. So we do see this as a very small market as far as players, kind of like the IGBT, if we look at it. So it'll work itself out, and we're going to be the leader in there.

Got it. On the idea -- you had a question?

Couple of questions, first one is from a business model standpoint [indiscernible]. Do you think that your direct relationship with OEM is going to be the way forward e.g. OEMs not going to in-source inverter but why do you think Tier 1 still have the major role today? And then related to that, I was particularly interested in contract you've signed with the TESCO. Not only a multiyear supply agreement, but also sort of $250 million in front to finance our CapEx. So do you think that could be the norm going forward where your customers are basically underwriting your company investments?

Yes. So on -- the first part of the question, the OEM, the answer is yes because the inverter is the main thing that is distinctive for how the vehicle performs, which is the identity of the OEM, right? The OEMs never outsources the engine today to somebody like just give me the engine that you give everybody else. It's a core differentiator. Inverter is going to be the same. Because the other thing is doing the right inverter for that platform will save on battery volume. If you have 10% more efficiency, you have 10% less volume of battery, that's a slam-dunk win for the OEM from a cost or extended range. . So the OEM is the core decision maker for the technology of the inverter. Whether they in-source it or not, that's dependent on the OEM and the scale they can. Not all OEMs can take that on. But from -- so but where we sit and the engagement we have directly with the OEM allows us to service both, meaning if the OEM does it, we deal with the OEM directly, or it could be a directed buy where we've -- the OEM already made the platform selection and the technology selection. And then they say, okay, we awarded to this Tier 1. Now you all work together to deliver. But the anchor decision is already made by the OEM. So that to me is how the business looks like. On the go-to-market, what I would call the go-to-market, look, we've talked about, I think, 3, 4 quarters ago where we've had co-investments with the customers. This is the main one we publicly stated an agreement with TESCO as a partner. But that model, we've been executing to already, and we've already had co-investments. And that's important. Look, we can afford the CapEx, right? But why is that important is the stickiness of it. Especially on the ones that outside of the TESCO, where we signed in the midst of the shortages and customers were contributing or co-investing with us, not necessarily on a -- then procuring the equipment, but co-investment in order to get a capacity reservation. What that does, as you go through these uncertainties, the customer is going to go, "Yes, well, I'm going to move more share to On Semi because I already invested in it. So I want to maximize the investment." So you get majority of the share if the demand goes down. And we've seen already that, that's why we always say we're shielded from a lot of the fluctuations because our customers value what we bring and have been moving in a win-win situation. We get the call 6 months before the demand is soft and say, "Hey, I see this. I may not need all of it. What else can we do?" And then we have that conversation, and we end up with a win-win, and that's what's been helping us navigate through these uncertain times.

And we've been building the capacity to support the LTSAs. So if you think about any new awards that we get require additional capacity. So that's why we've got customers co-investing with us to build out that capacity. These LTSAs are on average 5 years, but some go out 9, 10 years.

And that gives you a little bit insight on a secondary verdict on our technology. When a supplier like TESCO is a very credible at scale, even inverter EV supplier signing up for 10 years and investing. They've done the work about which technology is the best. You know what I mean? So that's credit to what we've been talking about as far as superior technology, efficiency and then the joint dye and package combination.

Maybe one thing, Hassane, outside of silicon carbide, how is the demand outlook for other parts of your automotive business?

Yes, I would say it's stable [indiscernible], I think we were saying earlier, if you look at where our Street numbers are for '23 and you exclude silicon carbide, you got the rest of the business, the base business and if you exclude the exits of $400 million, you got the base business that's growing kind of flat to low single digits. That feels good right now in an uncertain world. So that's why we keep saying it looks stable. We look at our LTSAs, we look at our reschedules and everything kind of feels good right there. So I would say the non silicon carbide automotive is a stable road right now.

On gross margin, right? It's a very strong journey. I mean we were supposed -- used to On at 36%. Now we are at like 46%, 47%.

36% now is the EBIT.

No, that's a great accomplishment. How is the journey from here to the long-term 53% goal that you said at Analyst Day?

Yes, there's 3 components, right? '23 is a transition year, we've been very clear on that. If you look at '22 for the year, we came in at 49.2% gross margin for the year. '23 is a transition year because we're ramping silicon carbide. That's got a headwind of 100 to 200 basis points. We brought in the EFK fab, and that's got a headwind as well. So we believe by the time we get to '24, we've got scale on silicon carbide that rolls off. But there's 3 components to get from let's call it, 49% to 53%. It's the ramping silicon carbide that is line of sight to get there. It is the favorable mix and there is 2 components of that. If you look at the growth rates we've outlined, it's more automotive and industrial, so you have a mix to a higher margin. And then you've got new products that come in, right? So we talked about new opportunities. We talked about growth opportunities and gate drivers and controllers. As that layers in, that helps drive that gross margin. And then the third component is the fab right, right? So it's optimizing the footprint that we have now. Getting the right mix of the right products where we get the optimal cost structure as well as monetizing the fab divestitures. So there is a $160 million of annualized cost that comes out as we exit those fabs. So we've divested and we still have production running through there. But as we bring that in-house or somewhere else, we actually save $160 million on an annualized basis. So you got fab right, silicon carbide and favorable mix that get us to that 53%. We don't think there's going to be a step function in that ramp. We think it's probably more of a linear kind of gradual improvement over time because obviously, the mix takes a little bit of time and the other parts just take time to monetize.

Anything on use of cash, like is M&A on the table at all? Or do you see that as a secondary part of the strategy right now?

Yes. Look, from -- you're right, we have a very good position to do whether it's M&A or our capital return policy in the order where M&A is secondary to our own business, if you will. And the reason I say this is it's definitely on the table. But what we've outlined at our Analyst Day for the next 5 years is not contingent upon a missing piece, meaning we have everything we need to deliver to our strategy and the 5-year plan that we've put together and the growth that we've outlined, we have all the pieces we need. We have the technology, nothing is lacking. What that means is we're going to be very disciplined in our M&A approach. We have the scale, so we don't need anything for scale. We have a lot of technologies. So we don't need to do -- What you're going to see us do is we will use our M&A and our firepower to do, for example, a technology tuck-in that will accelerate in R&D development that we already have or gets us to market faster or gives us a TAM expansion beyond where we are. So it has to be a strategic M&A in order to deliver better than we can based on what we've outlined, but nothing is missing today from what we've outlined for us to deliver. So it's like the best place to be in because that's where it can very disciplined, both on timing and value.

It's still auto industry, right? You're not making AI accelerators anytime soon?

No. No, but we -- for every all -- all the AI hype and all the AI accelerators, they all need power because they're all power hungry, which even puts more importance on the efficiency of the power ICs that Sudhir talked about in Analyst Day, 1% efficiency on that much power goes a long way from power consumption and sustainability, right, on the usage model. So we're definitely going to be -- are and will continue to participate from a power domain aspect of it.

Thank you, Hassane. Thank you, Thad.

Thank you.

Thank you.