NXP Semiconductors N.V. (NXPI) Earnings Call Transcript & Summary

All right. I think we're live. So good morning, everyone. I'm Chris Caso, Raymond James, semiconductor analyst. So welcome to the Raymond James Technology Conference. And with us this morning is NXP Semiconductors. With us from NXP is Kurt Sievers, CEO of NXP; and also Jeff Palmer, Senior Vice President of Investor Relations, someone whom with I go back a long way, Jeff.

Thank you, Chris.

We certainly do. So thanks both of you for joining this morning. Just a little bit of logistics. There should be a button to the right of the screen that you're watching now, which allows you to submit questions. So we will take questions during the session and those show up on my computer here, so I can read them out and ask the company. But so thank you for coming this morning.

And maybe to start out, what I wanted to arrange the discussion and kind of start out with some of the things that you spoke about at the recent Analyst Day because there was a lot of information provided, I'd encourage if you haven't seen it, if you're doing work on NXP, you take a look at the details from that Analyst Day, and what I thought was good with that is that as you laid out the growth projections into 2024, it wasn't open-ended. There was a lot of detail in there about where the growth was coming from, auto and not just auto, it's radar and domain processors, not just mobile, it's ultra-wideband. So I'm going to go through that in a bunch. But Kurt, maybe just to kind of start, there was a lot of optimism here of convincing investor, this is not just a cyclical improvement. This is -- they're structural, too. And maybe just kind of an open-ended question on what's giving you that confidence that this is more than just kind of a cyclical rebound. This is more structural, and it's more specific to NXP than just the broader industry.

Yes. Thanks much, Chris, and good morning, everybody, who is able to listen in this and see us this morning. Yes, Chris, indeed, what we clearly articulated in this Analyst Day, which is not even 4 weeks ago, is that we're going to almost double our growth rate going forward. So we come off a 5.5%-ish growth between 2018 and 2021. And what we said is that now we're going to grow 8% to 12% going forward, which really sits on a pretty nicely diversified set of growth drivers. So it's also not a one-trick pony only in 1 of our segments, but the 2 large segments being automotive and industrial and IoT; automotive, about 50% of the company, the industrial about 22% of the company, both will grow 9% to 14%. So really strong growth. And now why is this not just the cycle? What's going on? I believe it's really 2 things here. First, I think actually in those application spaces, markets are accelerating. Forget NXP for a short second. But it is really the markets which we are exposed to, be it electrification in automotive or be it the ADAS radar game in automotive, but also the connected edge solutions in industrial. These are markets which we've been focusing on for a couple of years already because we knew they would take off. It's really the content in these applications, which is shooting up. So there is a market momentum, which is carrying us. And that is not something which is cyclical, but it's really secular. But separately to that, we also outgrow those markets. So on top of that strong market momentum carrying us, clearly, and we will discuss more, I guess, in the next minutes, we have the design wins. We have the differentiated product and solution positions to outgrow that accelerated market even, and that actually delivers us than the 9% to 14%. Now in the environment which we currently are, I think we also have the third element, which I just want to mention here at the beginning, which is the proximity to our customers. I mean we've been suffering, I would say, from a pretty significant and challenging demand-supply imbalance. And what it clearly does is it has brought us much closer to our customers, often, the customers of our customers, which creates much more visibility and much more certainty with that into future design win momentum and also new opportunities. So in summary, this accelerated growth, Chris really has 3 legs. The 1 is the markets we are focusing on here are accelerating. Secondly, we are growing share in those dedicated and specific focus markets. And thirdly, I think the overall proximity and intimacy with our customers, given the current supply situation, is actually another proof point to actually propel this whole situation going forward.

Great. That's very helpful. And maybe as we dig into some of those specifics because -- what I've noted for you folks, and you're known for your exposure to the automotive market, almost half of the revenue. But when I take a look at the numbers you provided in the Analyst Day, there were 3 areas that were really more significant than others. And as I look through notes I've got here, radar, domain processing and electric vehicles, that's almost 40% of the growth in the company that you expect to happen between now and 2024. So with that, those are probably some 3 areas we should speak about now. And maybe starting -- well, -- maybe we'll start with radar. Maybe you can talk about what your position is in radar and why you expect radar to have such strong growth over the next couple of years?

Yes. Radar clearly is a very material element of this. And the whole background is actually -- it is a must-have technology in everything which is now labeled Level 2, Level 2+, eventually maybe Level 3 safer driving. Think about applications like the front radar, which is used to measure the distance to the next car, allows them cruise control, which is good for convenience but also for safety because automatic emergency braking systems, for example, which are mandated in several jurisdictions need radar as a technology. So there is a -- and that is not something which did fall from the sky yesterday, but I mean we've been working on radar for many years. And now there is really a breakthrough in terms of cars being featured with radar. So from a market environment perspective, I think it's very, very obvious how radar is now massively accelerating given the combination of safety and convenience features. There's also an element of convenience in there. If you think about, say, cruise control systems, which let you kind of self-drive on the highway but certainly, safety is the overriding argument. Now NXP's position in this is today a #1 leadership position with according to external sources, something around 40, 40-plus percent market share. And this is based on a number of pillars. One pillar is a system approach. We are not only doing the microprocessor for radar or the front end, the RF device, we have both. And it's not only that we have these 2, but we also complement it with the network interface and with the power management IC. So when we speak about radar and our very material position, it is a combination of the processor, the front end, the power management and the network connectivity, which is very relevant because it makes it for customers easier to use a one-stop shop solution rather than putting pieces together from different vendors. The second pillar for our strong position is the focus on 77 gigahertz. So radar systems initially in automotive were based on 24 gigahertz. And over time, the industry figured out that the precision of the measurement and the reach of the beam will be better if you use a higher frequency. So the industry is moving to 77 gigahertz. And there, we have an even stronger leadership position. So the more the industry is now transitioned from 24 gigahertz to 77 gigahertz systems, the more NXP is going to benefit because we have a much higher share position in 77 as compared to 24. Now if you peel the onion further within 77 gigahertz, then technology becomes a major differentiator. And technology means that initially systems were built on silicon germanium, which is actually a good technology. I mean it works, the trouble is only it is relatively, say, specific process technology, which you don't get from mainstream foundries. And secondly, it doesn't let you integrate with the processor because it's different to silicon. So there is no path for 1-chip integration and the typical race of getting smaller and higher performance, which we always do in semiconductors. Now NXP was the first one to get into mass production in automotive with 77 gigahertz radar in CMOS. So we made that step. We also have silicon germanium, but we made as a first one in a mass volume application, the step to CMOS. And since then, we lead the game here. So we are now in a position to have the integration path open and also the path to benefit actually from higher performance because we are from generation to generation, we move into smaller nodes in CMOS. We started on 40 nanometers, now moving to 28 nanometers. And you know the game in semiconductors that brings you cost advantages and it brings you performance advantages. So just zooming out again, radar is being picked up in many applications around the car. We have been a leader right from the start, but now really propelling this forward with the move to 77 gigahertz because of our CMOS leadership. And that we do with the system approach, which is not only the front end, but also the process of the power management and the network interface. So take all these pieces together, gives us currently a business which has a run rate of about $600 million. So it's very significant already from a size perspective this year. So we are not talking about future, but that's already what we have today. And going forward, we have at the Analyst Day, forecasted that we would grow this business with about 20% to 25% CAGR, which lands us at about $1.1 billion in 2024. So with that, it becomes indeed a very material element of the company. And I'm happy to say that 90 -- I dare to say actually, 90-plus percent of that revenue needed to grow from the 600 million to 1.1 billion is already designed in. So we have those designs. It's a matter of execution now to get the products out of the door over the next couple of years.

Right. And that's one of the things that's really notable in automotive because the design cycles are so long. The bad news is you have to wait for the revenue. The good news is you can actually see it coming, which is good for folks like us who are forecasting for living and make our job a little easier. And just as an aside, would you say that goes for most of the auto business? I mean, where you've got that 90% visibility into '24. That's most of the auto business, I guess?

I would say, for all 3 of them, which we also specifically spoke about, so the S32 microcontrollers or microprocessors for domain and solar computing as well as the electrification systems, I'd say typically between 80% and 95% is designed in. [indiscernible] imports depends on the run rates of these specific cars because it's really designed into specific car models. And of course, we depend in the end on the success of the sales of those specific models. Since it is many, many, many, I think in radar and Jeff, you have to correct me now, I think the radar, we set in 20 of the 20 car companies we had designed in, so it's very broad.

That's correct, Kurt.

Yes. So statistically, it works out very precisely in the end. So when I say we depend on the run rate of a specific car model, think about many car models because, again, we are across 20 OEMs, and they have different models. So in the end, the 20% to 25% CAGR in radar, it will work out pretty precisely given the broad statistical base we have here. And by the way, Chris, what I didn't say -- initially, I mentioned that we are in nice markets which are growing faster than the general industry, and we are outgrowing the markets. That's also the case in radar. So the 20% to 25% CAGR, we think is about 1.2x the radar market. So it's not only that we grow fast, but we even outgrow that particular segment.

Right. So that implies you would expect that 40% share that you have today in radar to expand and particularly, it makes sense to 77 gigahertz expands? Yes. Good.

Exactly. And in that -- sorry to dwell again on radar because it's a nice -- it's such a nice example. I mean this #1 position has already a good distance to the next guy. And with that, it's a good showcase of our obsession on relative market share. The whole idea is you know that we spent about 16% R&D across the company. If we spend now 16% of that $600 million, then in absolute dollars, that is more than any competitor can spend in that same segment. So the whole idea of a fast cycle in terms of more and more growth, better and better road map really works out given this true leadership position. So yes, the 40% should go to a higher number going forward, which makes our relative market share even better.

Okay. Let me pivot a bit to another aspect of automotive interest of time on electric vehicles and also something you spoke about. And there's a couple of areas in electric -- of contact electric vehicles. And I think what you identified as the battery management system was the most lucrative content there. So maybe you could talk a little bit about what you have in that space? And I'll follow on one of your competitors, Analog Devices has spoken a lot about this as well. So it's a good opportunity for you guys to position yourselves against them since they've been speaking [indiscernible].

Absolutely. I mean we -- first of all, the environment, obviously, with four electric drivetrains is wonderful because this is a very fast accelerated trend currently in the car industry, they move to electric -- fully electric, battery electric and hybrid electric vehicles. So that actually is going faster than we also had anticipated a few years ago. So the share of electric cars against the total is really growing much more rapidly than we thought. Now we talked about battery management systems, as you say. They are the key differentiator against the company like you just mentioned, like ADI, for example, is that we offer a system approach. It's a bit similar to what I spoke about in radar. So a battery management solution consists of analog front-end chips which sit on the -- on each cell in the battery. And then you have a microcontroller in the middle, and the microcontroller is actually steering all of these connected analog front-end chips in order to make sure that this battery is discharging at a very even pace that maximizes the power of the battery and with that, the range of the car. So it has a direct positive consumer benefit. And our differentiation is that we do not only have the analog front-end chips but also the microcontroller. And secondly, that we have the whole system under ASIL-D. ASIL-D is this cryptic thing, which is about functional safety, which is a very, very crucial feature in electric cars because a battery catching fire is obviously a bit of a problem, thinking about these batteries really constitute like a power. They are like a power plant on wheels. So when you think about them, competitors then we really leave them behind us now in growth speed because of, a, the capability of offering a fluid system solution, and b, offering that whole solution on ASIL-D. ADI and a couple of other competitors, which really come from the high-performance analog space, they don't have the automotive microcontroller as part of the solution. And that is actually why our battery management position has significantly improved. What we spoke about in the Analyst Day is actually not only battery management, by the way, but we also opened the kimono a little bit more and said we've also started to work on the e-motor control systems. And if you add the 2, then we are currently on something like a $200 million run rate in this e-car specific silicon, growing to 0.5 billion in 24 billion, and that's a 30% CAGR. So for those of you who followed us for a longer time, I mean this is faster and more than we had spoken about in the past. So the $200 million run rate this year is already actually more advanced than I think we had anticipated. And bringing this to $0.5 billion in total size by '24 makes it also by then certainly a more material part of the company. And a growth rate of 30% is also at least to our estimates, about 1.5x market. So if you then compare to other players and the total electrification market, we outgrow it by a factor of 1.5, which is quite significant.

Very helpful. I'll jump around to another section. As I try to compress your full day Analyst Day into 40 minutes. But one of the old ones industrial IoT, and I guess that would be the second largest growth driver for the company. And the CAGR for that at 11% a year between now and 2024, I think that one -- with auto, it's broken down into these elements. So it's a little easier to digest. But the 11% CAGR, what gives you confidence in that? And I guess it's harder to kind of point to one thing as you point to auto in order to support that growth rate.

Yes. It has similarities with auto, and it has a lot of dissimilarities. So that's why I totally understand your question. I mean, what is very different is it has a very long tail of customers. So that business is much broader-based from a customer perspective. And what it takes to be successful with that long tail of admittedly smaller customers, is our system solution approach. So the success here really goes to a solution cell. And at the Analyst Day, I think we also broke it down at least kind of directionally into what is microcontrollers and microprocessors in that and what comes from analog security and connectivity. The connectivity part, by the way, being the piece which we have acquired from Marvell 2 years ago. When I say solution, then here it is really typically this connected etch solution where small companies, which maybe have -- are building coffee machines or vacuum cleaners, but also big companies being industrial automation, but all of those people have not really been in the depth of electronics in the past. They were more electrical companies and now suddenly, they are exposed to a world where they have a connected application where cybersecurity is playing a major role, where artificial intelligence is coming in. They have these inference systems where think about the vacuum cleaner which does its job actually automatically in your home or the machine which is cutting your lawn, same thing. The manufacturers of these or the designers of these systems, they are suddenly exposed to artificial intelligence, cybersecurity, connectivity. And the only way for them to quickly get into this is what we are offering, which is reference designs, which is fully integrated software stacks, which makes their life so much easier. So I do believe the design win momentum, which we have, which lets us believe in this 9% to 14% growth is to a very large extent, built on the success of our system solutions, which if I think about our competitors, actually, there is not one company which has that complete set of -- from portfolio pieces which we can now put together into these reference designs. And admittedly, Chris, 2 years ago, we couldn't either. I mean there was the big reason why we were really obsessed to find an asset for connectivity because we felt this is a big deal for our customers, and they had to get it from somebody else, so the whole -- the tool chain for the connectivity piece was separate to the tool chain for the processor, very difficult. And now we've all unified this, and that's why we have really a lot of faith and trust in our growth here.

Got it. Very helpful. I'll take a pause here and just remind people if you want to ask a question, please put it through the portal. It's on the right side of your screen. So we haven't gotten any. I'm going to pivot here. We've got about 15 minutes left. And you can't escape a fireside chat without a short-term question or 2. But -- and it's really a prerequisite for folks doing work on the stock or any semis right now that people are worried that we're not overdoing it here in such exceptionally good times now. And what I'll bring up the auto production growth rate is, and it's still down as capacity constraints have hit the industry. But yet, the semiconductor shipments in the automotive are at a higher pace. And I know you've gotten this question a million times, and it's -- I'm just -- I think the audience would benefit from the answer.

Yes. So I think where it [indiscernible] down to is the question how can we grow 40-plus percent this year where the SAAR is. To the answer first, and then I'll try to argue why that is. So no, absolutely, we do not build any unwanted inventory. And we do understand, and that was not the case half a year ago. We do understand the delta, and the delta is made up by 4 points. One is the content growth, which has to do with one electrification, we talked about this. Secondly, with a strong shift this year to premium cars. So in a way to profitably survive, the car companies made all major moves this year to focus fully on their premium vehicles because they make more profit on those. So they were -- with the few chips they got, they were building those. But they have 2 and sometimes 2.5x the silicon content of a volume car. So the electrification trends. And secondly, the move to premium vehicles has massively driven up the content increases this year. The second element is our company-specific growth. So I am convinced, along with what we discussed in the past 20 minutes, we are growing market share. So there is an element on top of the content gains that we gain NXP-specific content. And then thirdly, there is an inventory effect, but it's not the one which you would normally expect. It is more, you have to go all the way back to 2018. And at the end of 2018, the biggest car market, which is China started to break down, in '19, it was crashing. And '20, we had COVID. And what has happened through all that period is that the extended supply chain, which is a very deep and long one between us and the car companies is just -- it has dried out to an extent which I would have never believed possible. So after 27 years or something in this industry, I've never seen it running so dry because nobody knew what the future would look like. So they all went to the absolute bare minimum. So we are now as an industry, and NXP is a significant part of it, we are busy to just bring that back to normalized levels. And I like that one example to illustrate this. We have customers which normally have a handling time of our product of 6, 7 weeks. So because they transport submodules to another place and then they do more manufacturing, et cetera. And given the enormous pressure, which is on the supply chain now, some of these people have actually condensed it down to less than a week. And they do this by immense profit losses, if you will, because they go to manual handling. They use air freights instead of ships, et cetera. They all want to go back to normal, and that means they need more products, but that is then product which is just needed to let the machine run, it doesn't help you to build a car more. It's just -- they have to fill the chain again in order to be operational again with -- in more normal ways. And that is also ongoing. So that's why I say we are not building inventory, but we -- there is a certain part of the product is going into the chain, and it is needed to just normalize it again and to go back to more normal operational levels. Now all of that is going to take, I think, the entire next year. We are fully ordered out for next year. I think on the earnings call, which is now 4 weeks back, we talked about the fact that I think 3/4 of our automotive portfolio has a lead time of 52 weeks. I mean, that gives you a feel of what's going on. We have NCNR orders out there. And we cannot even satisfy them for next year. So we don't have a 100% satisfaction rate of the NCNR orders, which are really non-cancelable. They cannot shift them until the end of next year. So all of these signals and maybe the last one is the distribution inventory, which is still down at 1.6 months for us, which is an all-time low, but we have it now already for a year like this. All of these elements taken together should give us the confidence that this is not an inventory or stock build event, but in the automotive industry, that product is needed, and it will take quite some time still until we are normal. And then, Chris, and that's just an outlook for the further future. I do believe, and I have that in my daily discussions with them, people want to build strategic inventory in the extended chain. I mean that they will want to do this, but nobody can do it at this point in time.

Great. And do you think it's even possible to do next year with what you're saying? You talked about being titled through next year, but that would get you back to normal levels, sort of 2019, 2018 levels, but not higher?

Yes. So -- and that's actually an element which makes us very confident about growth rates beyond '22 because then this would only get started.

Okay. I want to also ask you another question, another structural change that we've noticed in the industry. And a lot of this is spending time with GlobalFoundries that just did their IPO. So there's a lot of public conversation about that now. And it's about the relationships you have with your suppliers and obviously, that there's a need for more capacity. And these suppliers are requiring more long-term agreements, things we haven't seen in the past. And there's been a lot of discussion about what that means for the industry. Now that you have long-term agreement with your suppliers such as GlobalFoundries, your customers have agreements with you, what does that mean in the up-cycle and the down-cycle? Is that...

I actually believe, especially for markets like industrial and automotive, it is a very important and very relevant learning, which is now -- which starts to be applied, given the fact that those -- specifically those 2 segments have specific technology needs. So it is not like exactly the same technology which goes into mobile or goes into compute segments. And given the nature of our industry that you need a long time to invest before you build a factory and before you see your product, I mean, you easily -- it takes 3, 3.5 years. And in order to get this in the proper equilibrium, I think those longer-term deals and agreements are extremely meaningful, especially since they involve everybody, all the way down in that case to the car companies. I mean you've also seen public announcements with, I think GM had a press release, which mentioned us to be involved in deals. Actually, before Ford talked about GlobalFoundries, they had a press release of using our product across the board. So I think there is a trend now that the whole value chain is working out future production needs. And not only for a year out, but actually for a number of years out, which I think is absolutely required, and it will help to make this fully covered on the other side by NCNR orders from our customers, and I know they also get it from their customers. So the whole system is just more dried out.

Right. But -- and what implications will that have once demand eventually slows because we're still a cyclical industry, the forecasts at some point will be wrong. It's one of the concerns is all these non-cancelable orders, but they're based on a forecast that someone's provided that one day will be wrong. And it's probably not in your interest to put inventory to customers, they don't need in one day and your suppliers on you. So what will happen in that circumstance?

No, you're absolutely right. I mean -- and by the way, the cyclicality normally comes from building inventory. But that's a bad element of all of this. But with the transparency, which is now being built, which was not there in the past, Chris, I think -- and it will never be perfect, but I think we can avoid this to a much bigger extent than it has been in the past. Because the issue in the past was that you only -- in the value chain, you only spoke to your next level beneath. The tier one company would speak to the car company. Now this total circle is closed and everybody speaks to everybody, which I think will also give much earlier reaction times and faster reaction times to possible starting inventory builds. So I think that there is feedback loops. Now on the NCNR orders, you have a specific question, let me specifically answer that. I'm not worried about next year, Chris, because from what I said earlier, these commitments for the next 5 years, I mean we made forecasts together for a couple of years, but the commitments range into, say, next year. On the foundry side, we have some commitments which are longer, so for more -- for a longer period of time, but then it's actually small numbers. I mean you saw in our papers that we published a $4.4 billion commit to our supply base. But mind you that is not only foundries. And if that is over multiple years, it's not a big number at all.

Right. We've just got a couple of minutes left. One of the other things that's different structurally with the industry is what's going on with pricing also. And ours is not an industry where typically we see annual price increases. I think we're all comfortable with the reason why that's happened this year, right? The cost of lagging edge manufacturing has gone up, its new tools and new capacity needed. But I guess the question is, is this something structurally we can couple of percentage of price reduction per year, is that going to change now because of what's going on with the cost structure of the industry?

Yes. So first of all, let me just clarify again what we are doing. We are increasing prices, absolutely right. But we increased them in line with the cost -- with the input cost increases, which we are exposed to. And we do it to the extent to completely protect our gross margin percentage. So we don't want to [indiscernible] our margins with that because given the long-term nature of relationships in those industries we are exposed to and the products which are not commodities, but application specific, I believe that's the right thing to do it. But we absolutely make sure that we will not drop gross margin because of the higher input cost. I do believe that the cost increase is a reset of the industry. So I don't think this is all going to drop down again in future. But of course, at some point in the future, we will see, again, annual, say, ASP decrease, maybe to a lesser extent than in the past, but it will not fall all the way back down to where we came from. I mean we do know -- we make a move up. And I think from that move, we have to see if it needs even more in the future, and then maybe there will be, again, slowly ASP erosions depending on the product, depending on the business, on the geography, et cetera. But the point is, I think we are seeing a reset as of now given the investments which had to be made on both our side but also especially on the foundry side. And I -- I mean if you take automotive because you were asking about the growth and how is it there, I mean if you think about the fact that average car has a $400 million to $500 million semi content, I say only. Now imagine the price for that goes up by whatever, think about 5%, 10%, think about 20%, that is not much to a car. I mean, if a car sells for $40,000 and you increase that $400 thing with 5%, 10%, 15%, 20%, it isn't actually a big deal. And that's why I also think it will be swallowed, it will be absorbed, and it will not be an issue going forward.

Got it. All right. That's very helpful. I think we're at the end of our time. So I think we'll leave it there. Kurt, Jeff, thanks for joining us. And thank you.

Chris, thank you very much and everybody on the line. Thank you.

Thanks, bye-bye.