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

Okay. Okay. Great. Okay. I think we're going to get started. We'll jump right into it. Welcome. Very excited to host NXP for the third of 5 straight semiconductor fireside chats. Honored to have Kurt Sievers and Bill Betz. Kurt has been in NXP for 27 years. He has served as President and CEO since May of 2020. He's been a member of NXP's executive management team since 2009 and prior to being elected as CEO, he was the President of NXP and oversaw all the company's business lines, played a key role in the merger of NXP and Freescale Semiconductor in 2015. Bill is an Executive Vice President and CFO at NXP. He's been there since 2013, a 20-year veteran of the industry. And before NXP, he had several financial leadership positions with Fairchild, LSI Logic and Agere Systems. So lots of great industry expertise here. So Kurt and Bill, welcome, and thanks for joining us today.

Thank you, Mark.

All right. So if you don't mind, I'd just like to jump in and start off with kind of a review of the strategy of the company. Over the last 10 years, there's been a lot of consolidation in the microcontroller industry, including your own acquisition of Freescale. Now if you look at the microcontroller industry, there are 6 companies that account for 80% of the industry. Many investors consider this an elite group of highly profitable companies. Healthy growth, exposure to the right end markets and secular drivers like automotive. How should investors think about NXP relative to the other high-quality companies in this elite group?

es. Thanks, Mark. So first of all, I would agree our aspiration is to be the leader in secure edge computing, which is about a system approach, but the core of the system is a microcontroller and/or a microprocessor. So agreed. Now I'm also happy to say that we are actually, at least according to Gartner, we are the #1 when it comes to microcontrollers across all sectors. So if you just take the total global microcontroller market, we turned out as the #1 leader last year. And one subsegment of that, which is very relevant for us, obviously, is automotive, where we are also happily the #1. So I'd say, yes, it's a strong group, and there is a strong leader of that group, which is NXP. Now from a strategic perspective, I think the main growth in the semi industry for the next probably 8 to 10 years is in automotive and industrial electronics. Those 2 sectors are really outgrowing wireless compute and other segments. And all of these applications have, as the core, a microcontroller or a microprocessor. When I say or boundaries are blurring. So we have this concept of crossover processes where you need the real-time capability of microcontrollers, but you need more and more also of the feature set of application processor. And that's something where we have a strong leadership in both areas, so we can successfully get it together. So I would agree, this is a core part of the market. It is a core part of where the growth opportunity is going forward. And I think especially after the acquisition of Freescale, NXP is in a very strong position there.

And one, I'll just pick up on that idea on Freescale. You were a key part of that decision and integration in 2015. So we're 7 years beyond that. What was the rationale? And how did Freescale improve your competitive position?

Yes, it was really exactly about that. So first of all, now in hindsight, it's good to say it was both a transformational deal for NXP and a successful deal. I think we're 6 years into it. We can say that one worked. The rationale was that NXP had a strong leadership before that in security, in parts of RF and in analog mixed signal, but we were actually very weak in the embedded space. And that is the main part which we were aspiring to get through the acquisition of Freescale, with a vision going forward to be more of a systems solution provider to our customers, putting actually solutions reference designs together, which go all the way from connectivity, sensing, processing, over actuation, and we were missing that -- especially at that time, that processing piece, and that's what we got through Freescale. Now admittedly, that hasn't totally completed our vision and our aspiration because we were still missing connectivity. If you think about these connected edge solutions, it always needs connectivity. Now we know we had a bit of an adventure with Qualcomm for a little while, which would have satisfied the need for connectivity that fall apart in summer '18. So by the end of '19, we closed the deal to acquire the connectivity assets from Marvell, which is on top of the Freescale acquisition, actually, that piece, which made us now, I think, an unrivaled leader when it comes to completeness of edge compute solutions.

Got you. I'm going to shift gears to the operating model, if that's okay with you. So after you divested your discrete business Nexperia in 2017, your gross margins were in the 53% range. They took a dip a bit when COVID hit like many other companies. And then since then, they've jumped 500 basis points last year to 56%. You just reported 57.6% gross margins in the most recent quarter. And at your Analyst Day, you guided for your 3-year gross margin target to be in the 55% to 58%, right? So you obviously believe that the increase in the gross margins is permanent. What were the primary drivers of that 400 to 500 basis point jump in the margins before COVID?

Sure, Mark, let me take that one. Related to gross margins, which is getting us to 58% is the internal utilization of our factories makes up about 40% of our front end. And we're running those in the high 90s. So that 500 bps, a lot of it was driven compared to COVID days when we were running around 60% utilization. So that was a big step up. Then I would say the higher volumes naturally with the fall-through associated with that. And then lastly, product mix continues to play a big important role at 58% and what gets us above 58% in the outer years.

Got you. And what we hear from a lot of semiconductor companies is they've had to increase pricing to offset the higher input costs. Can you help us understand, is that part of what you guys have had to do? Is that part of the gross margin story also? Is that something completely separate?

So we have a clear policy at NXP. We only pass on the higher input costs to our customers and to maintain our margins. We are not here to pad our margins in any way we can. We think being transparent to our customers on this helps us in the long run and will pay dividends in the future as we don't take advantage of our customers during these unfortunately high inflation times.

Got you. And as you're approaching the upper limit of your target range and perhaps you got there, I don't know if you got there more quickly than you expected. Is there a physical limitation to your gross margins? Or can you say them going higher as you're adding more value to your customers?

Well, there's no physical limitation to gross margins. Obviously, they can go higher, right? It all depends on the portfolio that you have at hand. And over the last couple of years, we have improved that portfolio that got us to this respectable gross margin figure around the 58%. Again, in any given quarter, maybe 50 bps can move the gross margin plus or minus, but what really gets us at 58% and beyond, it will take several years because the investments that we are making today, we're very disciplined from the markets we want to play in, we want to be at a relative market share of 1.5 greater than our peers. And then more importantly, we have entitlement models, very financial discipline on our returns to be at our corporate margins or above in all our new investments. So it will take us time as we mix up the portfolio over the long run to sustain the 58% and go beyond.

Got you. The industry has been in a supply-constrained environment for a while. Do you have a view why did we get to this point of being in such a constraint? And is it changing foundationally like how you guys run your operations either with your suppliers or your customers?

I think we've learned a lot on why we are where we are. And I believe at least from the markets we are serving, the main reason is that it is about trailing edge technologies, so 28 nanometers, 40, 55, 65, 90. So that bracket is the bulk of the volume, which we supply. And I believe after a difficult 2019 and a very difficult first half of 2020, actually, I think the industry has structurally underinvested into these trailing edge nodes relative to the explosion of demand, which is driven by all the good things which are really secular drivers like electrification in cars, and I guess we speak about this later. So I would say you have, of course, all sorts of variations and modulations on top of that with the supply chain issues with fires in factories and winter storms and COVID shutdowns, but fundamentally underneath, there is an underinvestment in factory capacity in semiconductors for trailing edge nodes. That, I think, is what got us into this. Now the good news is there is a solution to it because, I mean, we can catch up. It's going to take time. But to me, it's relatively straightforward what the issue is and what can be done about it.

And until you catch up or till the industry catch up, do you have to change your business processes? What has changed since things have gotten tighter?

Yes. What has changed, and I think that's here to stay because it's a positive for everybody in the value chain is a much higher level of transparency on future needs and demands, not only with our direct customers, but with the customers of our customers. Think about big industrial customers, which are served through distribution, think about car companies, which are served through Tier 1s. We have now, on the highest level, access directly to them, get into supply chain discussions but that has morphed into much more strategic discussions, which are about innovation in the first place and then supply assurance along that innovation for many years to come. So what it means in the end is that we will have transparency about the demands for the next 5 to 10 years. It sounds long, but in automotive, that's what it takes. And that goes all the way back to the foundry partners. I think that clarity is helping to plan everybody in the process. It would not have happened if there hadn't been this painful period of supply challenges or shortages over the last 1.5 years. So that's a positive outcome.

Got you. And so this is something that's been in place for the last 18 months or so at least. And then there's a new kind of challenge that has arisen, and that has been more concerns about inflation and higher interest rates and potentially the risk of a recession. How -- is there a framework that you have to think about what happens to your business in either a higher interest rate environment or a higher inflation environment? Is there an impact to your model? Or how you're dealing with your customers?

Well, there is 2 sides to this. The one is the external side. I think we -- by the longer perspective and time horizon, which we do planning together, and I mean nobody has the glass ball, who knows what exactly it's going to be in 7 years. But by having that dialogue on an ongoing basis, I think we are much more alert to when and if there would be changes. From a more internal perspective, and I know Bill will give you a bit more detail, we think with our hybrid manufacturing model, we actually have a good mix, which means we don't have that brutal problem of underloading factories when demand would drop. So we have a relatively low portion of fixed cost. We've shown that nicely, I think, Bill, in the first half of 2020 when the world broke together through COVID where we still printed strong cash flows, but maybe you'll give a bit more detail on that.

Yes. No, absolutely. Just to add to that, Kurt, I'd say the other metric that we are very paranoid about is, as you all know, 55% of our business goes through the channel, our distribution channel. And unfortunately, it's been stubbornly low at 1.6 and 1.5, and we want to get it back to 2.5 during normal times. It's difficult. And I can assure you internally, we look at this daily, weekly. In any place where inventory is building up if we get a sniff of it, we have thousands of escalations from our customers in other parts of our business that we work with this customer care of redirecting that material on a weekly basis because our #1 priority is to prevent lines down for our customers. I'd say that's half of our business related to it. And then the macro economics, again, we watch this very carefully like everyone in this room. Again, with a PMI above 50 is still a good outcome for the semi industry. Clearly, that metric correlates very nicely to our business. So we watch that very carefully from an industrial standpoint. And there's many, many other metrics that we constantly measure and understand if there's any reflection points on those metrics. Until now, in the markets we're serving, supply is here, demand is here. Even if we cut demand for double ordering, even if inflation comes, we -- demand again, it's still going to take us time to get to that equilibrium. As Kurt mentioned, with the lack of capacity in the areas of our mature technologies that we're ramping just now.

Great. That's helpful. Let's shift gears to the -- to some of your businesses. So automotive, about half of your revenues. And at your Analyst Day, you argued it would be the highest growth segment along with industrial IoT. What are -- you have previously shared a framework to think about like the different segments in auto. What are the fastest-growing parts of that business?

Yes. I think that's an important concept to understand that indeed, almost 75% of the company are automotive and industrial. Both are actually growing at 9% to 14%. So I will focus first on automotive, but it's important to understand that also the industrial side of the house, growth at the same speed, 9% to 14%. And in brackets, I would add the -- we made those numbers and gave those numbers before the whole effect of price increases was visible to us. So we did this at the Investor Day in November. So those 9% to 14% are actually ahead of pricing. So there is a positive offset to that, thanks to the pricing going forward. Within automotive, the biggest one from the fast growth drivers is radar, which was a $600 million business last year, growing to $1.2 billion, which is a 25% CAGR until '24, so between '21 and '24. We are the #1 and -- now we only gave the 3-year guidance. This is going to grow much lower because the penetration of radar, the multitude of radar sensors per car and the increasing value per sensor, which is because of imaging radar and higher performance of these radars, is ever growing. So I think the '21 to '24 is kind of the early innings and there is more to come, but it already becomes a very sizable business for us with $1.2 billion in '24. The second -- so that falls into the trend of safety and automation of driving. The second one is battery management and inverter control, which clearly falls into the electrification, which is an enormous content boost currently. There, we're going to grow to about $500 million coming from $200 million. So $200 million last year, $500 million in '24, which falls actually nicely like radar into what I explained earlier about the combination of Freescale and NXP. In both cases, we have a system approach. So in BMS, we are offering the high-precision analog front-end chips on each of the battery cells and the microcontroller, which is steering the system. And we are the only provider which has that complete system approach, and it is thanks to the combination of Freescale and NXP. Same, by the way, in radar, where it's about front-end transceiver chip, 77 gigahertz in CMOS and processor chips, which, again, we don't really have a competitor which has that strength on both legs. The third one in automotive is the domain compute, which -- that's a bigger discussion, but in essence, the compute architecture in the car is significantly changing from a very flat, simplistic network architecture to more concentrated compute power in domains and zones. And we are leading as the #1 processor supplier to the auto industry. We are leading that transformation. So domain computers are also going to grow to like $500 million, $600 million in the next 3 years, which again is very early. There is much more to come after that. I'm very proud, Mark, that just I think it's now 3 weeks ago at Computex in Taiwan, I showed to the world the first 5-nanometer test chip from NXP fully ASIL-D compliance, so that's a fully functionally safe automotive 5-nanometer processor, which is without competition so far. I mean, there's just nobody who is that far advanced in going to leading edge for automotive processing.

Got you. And the -- and when we think about your auto business, between 2017 and 2020, it was kind of in this $1.1 billion per quarter range. Last quarter, you did $1.6 billion. A lot of times, semiconductor investors get nervous when you go so far above the trend line. So quickly, what happened to get you so high above the 4-year run rate?

Well, I think the nervousness can be dampened if you draw the trend line all the way from '18 to '22 because what happened is, a, in '19, China car production came down and the whole extended supply chain depleted inventories brutally. So we were under growing the SAAR. That's not because we lost sockets, but they just depleted inventory. And then early 2020, the COVID shutdowns happened so that even got more extreme. And since then, it came up again. So each single year doesn't make sense to look at. You have to look at the whole time. And then actually, it works out. If you think about the content increases, which are going very fast, we have today, I think, 23% of the SAAR is electric vehicles. And they have 2x the content of a standard vehicle. Secondly, market share gains. I mean, the 3 areas I mentioned earlier, battery management, radar, domain compute, we are winning share. If you factor these elements in plus more recently pricing, it's actually not a reason to be nervous. It's actually a reason for confidence that we can continue on that growth curve.

Got you. And you talked to a lot of investors, what do you -- do you think there's a part of the automotive business where they are going to be really surprised in the next 5 years? Is there a part that investors are not tuning into?

Yes, I think they will be surprised about the rise of domain computing, where we make major leaps forward, both from an innovation perspective and actually offering products, which are unseen off. And a lot of this goes, by the way, in the idea of facilitating software updates into the car. The whole idea that the car is connected to the cloud, you get services into the car, and you have to do that with full cybersecurity protection as well as the right speed in facilitating it inside the car. Now the good news, by the way, for all of us as consumers is that the performance of your car will become upgradable. You are not locked down to the car as you bought it, but it will become better over time. But what it needs is headroom on the hardware side at the moment the car is built because the hardware is fixed. And that is, for us, a major upside opportunity. So I do believe investors will be quite stunned to see and we hope also to do some PR about this pretty soon, what is going to happen there. Finally, the model to do this is more and more moving into us doing it directly with OEMs. And that has to do with the whole supply situation I explained earlier. So rather than driving this innovation with 12 different Tier 1s, we do it directly with the OEMs, and they will play it back into the Tier 1s, which is actually, for us, a more efficient model to drive innovation in business.

Thanks for that. And I thought you were going to say ultra-wideband. Where is ultra-wideband in the development cycle?

Well, ultra-wideband is -- and I just saw one down here actually in the Street. If you take the IX BMW, which is this higher-end electric SUV, which BMW just launched, you can open, start this car with ultra-wideband, actually with NXP's ultra-wideband. Same for the Hyundai Genesis. So what I tried to say is across iOS and Android, you have now ultra-wideband going into automotive, that's the automotive angle. Of course, it is penetrating in mobile. And we are confident to also see now the penetration coming in IoT, which is tax -- smart tax to -- for lost and found kind of applications, open the door of your home. So the whole idea is to, like we moved the credit card in the past into mobile phones for mobile wallet, we move now the keys into your phone. So you would get rid of mechanical keys and have a more convenient and more secure system. So it's well underway. I think our unique differentiator is that we built that ecosystem on the security, on the ultra-wideband radio and on software, just like we did with NFC and Secure Element, actually almost 10 years ago. I mean these things take time, but we repeat now the same success in ultra-wideband.

Got you. And maybe you could just spend a little bit of time on in the industrial IoT market. What's the most exciting opportunity? What's your -- what's the secret sauce?

Yes, the secret source of NXP is, I believe, system solutions. There is a long tail of small customers that's different to mobile and automotive. Think about companies building vacuum cleaners for the smart home or connected coffee machines at Starbucks. Those companies often do not have the electronic design capability to do cybersecurity-protected cloud connections. So our added value is to offer a complete system solution and actually help them over the hump of the R&D, but also the time to market challenges. And that is exactly leaning back to what I said earlier about the system solution capability, which we architected over the different acquisitions, which NXP did. So with that, we see major success now through the channel to serve all of these small customers, which are making up industrial automation and smart home markets.

Got you. Okay. I'm going to -- we have about 3 minutes left. I do want to hit you with a cycle question. Your lead times as well as those of your peers have been stretched to the highest level in a long time. You were actually able to increase your days of inventory by about a week year-over-year and sequentially. So is higher inventories on your balance sheet, is able -- are you able to translate that into shorter lead times to your customers? Or at what point would you expect...

Yes. So first of all, indeed, we are with 80% of our portfolio, we are at 52 weeks or more lead time, which is horrible, if you think about it. The DIO had increased by 6 days. The only reason and I'm glad we could do this was to prepare for the growth in the second quarter. I mean we guided for sequential growth from Q1 to Q2 and in order to facilitate that we needed more DIO. So that's -- and that's it. So no, it -- that is not yet helping us on the lead time on the total. So I think and I know there is all the concern around the macro, but we continue to be sold out for the rest of this year. And given the structural situation on the trailing edge, I also personally believe that next year continues to be tight. Now if there is a brutal global recession, where, say, the SAAR would drop down to 65 million cars only, and mind you, that stacks against 95 million a few years ago. I mean, that would be a very significant drop. Of course, that would make a difference to our demand pattern. And I mean that's not something we can control. We are super alert. But from all the data points and Bill gave you a good overview of what we are tracking like channel inventory, et cetera, we can't see it yet. But again, I mean, we are not immune to this. We look at it very carefully. But currently, lead times continue to be long. We continue to be sold out. I have a long list of customers who absolutely want to sign up for a full year of supply of next year, sign the order tomorrow for the whole of next year, which we cannot serve because we don't have the supply.

Got you. Well, we have a minute left. I got about 20 more questions. So we're going to go speed round. Would you like to wrap up with any other final thoughts?

No, I just want to say, I believe, from a longer-term perspective, we are very well positioned given our exposure to auto and industrial, which are the fastest-growing markets going forward. Short term, clearly, environment is becoming very volatile and uncertain with all the horrible things happening around us, but that shouldn't derail us from long-term confidence in where we are going. And I believe there is a significant challenge for the whole industry in trailing edge.

Great. Kurt, Bill, thank you so much for joining us today and at the conference. Really appreciate your participation and your support.

Thank you, Mark.

Thank you, all.

Thank you.