QUALCOMM Incorporated (QCOM) Earnings Call Transcript & Summary

Good afternoon. We'll get started with the next presentation here. The next company is Qualcomm. And we have with us Patrick Little, who runs the Automotive Group there. So what I'll do is -- we also have Mauricio and Alexis from the Investor Relations group in the audience in case you want to grab them after the presentation. But let me start by just asking Patrick to give us a recap of the product announcements you had yesterday and also maybe introduce -- kind of just give a bit of background about your role as well as how many years you've been at Qualcomm and all that.

Okay. You have to be careful because I can monologue, so if I -- firstly, thanks for having me here. A pleasure to see many familiar faces again. I've been coming to CES for more years than I want to recall or mention. But I think last few years, particularly this year, it's really felt like automotive has caught a lot of wind in its sails. There's just so much going on in the automotive industry in terms of innovation. And it's been a century in coming, but there's -- and I think Mary Barra said it at GM, I've said it many times, there's more innovation coming in automotive in the next decade than there has been in the last 120 years. It's just we're hitting this knee of the curve where that whole use model in automotive is changing so much. And so I'll go through a very brief history lesson at Qualcomm just so you know where we are in the fairway, and then I'll talk a little bit about -- briefly about some of the announcements we made yesterday and what those mean to the industry. Qualcomm started about 15-ish years ago in automotive in the telematics category. And so it was really working with GM OnStar in 2002 that really got us our start. And back then, we were trying to define real value for the consumer around safety. And so that relationship with GM, one of the big announcements this week, frankly, was -- has really evolved over the last 15 years. And one of the things we learned at Qualcomm is this isn't one of the industries where you can go it alone. It's not one that you can germinate in a petri dish in a lab. It's one that you have to be working with the point of effect, which is the automakers. And so for the last 15 years or so, we've been shipping modems into cars. And honestly, as recent as 5 years ago, most automakers didn't even know why they're putting the modems in the car. They just felt like, "Hey, it's probably a good idea. Let's connect the car to the world around it." Well, they all figured it out now. And so modem attach rates have gone from almost nothing to 25% to 45% this past year. And this year, we'll tip over the point where more cars will be shipped with modems than without. And then by 2024, we're looking at 75%, 80% penetration of modems in the cars. And really what's going on there is simple things like diagnostics and being able to have the automaker connect with not the car -- really the car, but more of the consumer in the car has obviously been a point of value -- delivering value. And so now you really see many automakers like Ford saying that every one of their cars globally will be connected over the next few years. That's a very big business for us, fun business, frankly, and very adjacent to what we're doing in the mobile -- in many of our markets, mobile, IoT. And it's one of the very heavily leveraged business that helps make automotive a very accretive business in whole at Qualcomm. About 5 or 6 years ago, we started to look at our assets, which we now do on a quarterly basis, and say our compute assets that we offer in mobility can actually be a really good match to what we could do in automotive. And so we started to dabble a little bit in the infotainment space. And it was just then trying to figure out what it wanted to be because in years past, it was a little radio with a little twisting knob, which is a dinosaur now of the past. And over the last 5 years, with several customers, GM one of them, Jaguar Land Rover, another announcement we made this week, I used to make it my goal to be the trusted adviser to the automakers, be the one that they can trust as a technology partner to come to. And that's still my goal today, but now what it's really turned into is a cross-learning environment where they're teaching us and we're teaching them. We bring the technology, we bring the assets from GNS to modems to multiplicity of compute cores and then they bring the real idea of what they want to do with the user experience. But they don't want the user experience to be as you buying a car and then 4 years later, when that car breaks down, maybe coming back to buy another car. They want a relationship with you. They want to be able to have the relationship you have with your smartphone. They want you to bring that digital life into the car. And so they -- and that's, I think, one of the main reasons why they reached out to Qualcomm because that's fundamentally what we do. So the infotainment business, which was just a gleam in our eye some 5 years ago, has really turned out to be -- honestly, it's really shocked us in terms of our success and our projections on the yield from that business. Again, highly leveraged from our mobile business. In the mobile market, we sell application processors that have a variety of cores, GPUs, CPUs, DSPs, and our goal is to make that phone work for a day, 2 days, 3 days. And so we took that same experience into the automotive market saying we think that we can help the automakers pull a lot of cost out of their vehicles by helping them to integrate it. There's typically tens of electronic control units, ECUs, across a car. We really felt that our sweet spot starting about 5 years ago is to really help them to integrate all those ECUs into a few domain controllers. And boy, were we right. That really hit the sweet spot at just the right time and that's fueled a vertical growth in the automotive business at Qualcomm over the last -- particularly over the last 4 -- 3 to 4 years. So telematics, putting the modems in the cars, connecting it to the world around it, which will become increasingly important and sophisticated as we move to 5G. And then becoming that brain, that compute center inside of the car has also become a real sweet spot for the company that's going to be yielding dividends for decades, really decades, and it continues to evolve at a tremendous pace. This last -- yesterday actually, at our press conference, we decided to come out with a press announcement around an innovation that has been going on for quite some time at Qualcomm. It's been the worst kept secret at the company for at least the last 5 years. And that is that we launched our Qualcomm Snapdragon Ride autonomous driving platform. And this is something, as I mentioned, we've been working on AI for the last 5 to 7 years in our corporate development group for the benefit of all of our businesses, but really with a focus on automotive, how we could bring AI to the autonomous driving equation. And my team asked me from year after year, is this the year to announce? And we really wanted to wait until we found that intersection of value where our technology could solve the real problems that the automakers were experiencing, not Level 5 immediately, which we all knew wouldn't happen immediately, but how do we progress from Level 1 to Level 2 to Level 3, all the way to Level 5 smartly over the coming years, progressively and smartly, like all of us know, kind of knew would happen. And so we announced, I think, a distinctive offering in a couple of ways. The first way is that it is a configurable system. So it is the entire system, hardware, software, algorithms, the entire thing. We felt that throwing a chip or the software over the fence was going to be without any value at all. We felt like what we were trying -- automakers would come to us with use models that they've tried to satisfy like highway autopilot driving or stop-and-go traffic or lane keeping, and then many of -- we feel our competitors would give them features back. They would say, okay, so many tops or so many watts. That doesn't solve the problem. It just further complicates the issue. So what we did instead of announcing products over the last few years, we really focused on working with automakers and the one that we announced that we've really had a serious development partnership with is GM, not only in telematics and infotainment, but also in ADAS. And together, we've really been able to create what we think is value that is meaningful and can be commercialized more readily and more importantly, more scalable over the years. And so we will work along with GM and a variety of other partners that will be announced over the fullness of time, and how we're going to move from Level 1 to Level 2 to Level 3, all the way to Level 5 smartly over the coming years and risk reduce it where consumers will understand the value, where the systematic cost and risk will be right, where the power budget won't be 700 watts. It would be more like 100 or 130 watts, even a Level 5 system, so bringing the power budget down to reality because it's going to be in the back of an EV. Imagine that this -- your supercomputer brain for autonomous driving can't be burning 700 or 1,000 watts in the trunk because that's going to cut your range down seriously. And so one of the challenges that GM and others put in front of us is you have to solve the equation in the right parameters. You have to solve it in a way where it can be commercializable, where the thermals and the power and the systematic costs and everything can make sense. We could actually roll this out in broader volume and scale this out to our customers. And that's what we're announcing -- yet what we announced yesterday. Snapdragon Ride platform, we think, is distinguished from all the other offerings. I get asked from time to time, including from our own CEO, is the timing right? Was the timing right to announce? And I feel like this is one of these projects where quality means everything, getting the right spec at the right time. I didn't feel the need to get out there quickly and announce something that was a supercomputer that wasn't really seriously solving the problem. Well, 2020 has come. I feel it's the right time to start to deliver this value and we're working with automakers. You'll see the first instantiations of this go out into the marketplace in 2023. We're already working with several automakers. We already talked about GM, but there are quite a few that are probably obvious if you look through who our current customers are. 19 of the top 25 automakers are now -- we're now incumbents on the compute side at those companies. So that could give you some idea about who are the others that we might be working with on what we feel is probably the only practical, realizable autonomous driving platform out there.

Good. Patrick, let me try and kind of pull all that together and see if we can quantify some of those opportunities. At the recent Analyst Day, Qualcomm talked about a $4 billion SAM opportunity in automotive by fiscal 2022, which I assume then the ADAS opportunity is kind of more fiscal '23 in terms of seeing some revenue out of that. As we look at then kind of this $4 billion opportunity that you're looking at, can you kind of break that down in terms of how you're thinking about the rest of the opportunities that you talked about and the magnitude that each of them contribute to that $4 billion? And then maybe as a next step, take a longer-term kind of view in terms of then as the ADAS opportunity ramps, how does that look maybe another 5 years from then?

Yes. Great question. So as you sit here today, we have about $2 billion worth of opportunity in front of us. And that's really focusing on, I think, 2 businesses. One is what we call connectivity, which is both the modems and also Bluetooth, WiFi type technologies in the vehicle, and about half of that $2 billion opportunity is in that category. The other half of the $2 billion opportunity, I'm even looking slightly backwards into 2019 to give you a static point to kind of reference, is really the compute in the vehicle, which really I define as infotainment solutions. So half of that $2 billion is kind of the connectivity realm of our business and half of that is the compute realm of our business. As you look forward to 2022, which is only 2 years away, that SAM will double for us. Some of it has -- a lot of it has to do with our -- looking at our assets and looking at the opportunity in the automotive industry and realizing that there's a whole lot more that we do already in the company that is quite applicable into the automotive space. And primarily, I think, one of the businesses I should punctuate is our RF front end business. So particularly as we move to 5G in automotive, an example is this. We're the kind of the industry incumbent in 3G and 4G telematics. 5G, we've come out very strong, very early with a unified platform, and we have not had one customer design win on 5G that did not include our RF front end solution. And so that's a really -- 5G is more complex. You really need a tight coupling of the baseband and the RF front end. And so on our 5G solutions, we've coupled our RF front end solutions with that. And that is -- a big part of that growth, that SAM growth from $2 billion to $4 billion is the inclusion of additional assets like RF front end, that's really going -- is a new category for us. I'll look back even a year and there was scarcely any RF front end from Qualcomm in our automotive designs. When I look forward, I'd see a majority of our designs, including our RF front end components along with it. That's -- in addition to that, you're really starting to see the inflection of the curve in the connected car. And so as I mentioned, this past year, let's call it, 45% of the cars rolled off the line connected with a modem. As we move into this year, it will easily go past 50%, well on its way to a nonlinear path, about 75% of cars being attached by 2024, which I think is an incredibly conservative estimate. And the reason for it is this, the automakers want that user experience. They want that relationship with you. They -- the days of them just delivering you a product are pretty much over. They want to deliver a user experience to you. And they feel like services are a big part of that user experience. And so one of the things that we announced this week and one of the things they came to us as a trusted adviser and said is, "You helped us deliver bits to the car. Can you help us to enable the services that will go through that pipe?" And so one of the things we're doing and specifically is over-the-air updates, allowing the car to update itself. So instead of pulling into a shop, of course, you've heard of over-the-air updates, being able to repair a car on-the-fly or overnight in your garage. Now in addition to that, what most of them want now is over-the-air upgrades. They want the ability to be able to offer packages, just like you have on your handset, just like you have in your home entertainment in your house to say, "Oh, I'd like a surveillance mode on my car. I'd like to be able to tap into the ultrasonic sensors and the cameras in my car. And since I'm in Las Vegas today and my car's in San Francisco, my ultrasonic sensors just woke up and told me that something was immediately on my car. And so I want to be able to see what's going on, on and around my car. Is it a break-in or a pigeon?" And so that's an example of -- probably a pigeon, it was San Francisco. So that's an example or driver monitoring for medical and for fitness or wellness purposes. So delivering services is going to be a very important part of the hopeful relationship between the automakers and the consumers over the coming years. And that's also a part of that growing SAM over the coming years. Now as we look past that horizon, I think the real growth is in ADAS. The ADAS SAM that we already calculate based on already the assets that we're intending to employ in this space will double the SAM for us. It will add another $4 billion to $5 billion worth of SAM starting about late 2023 when we first ship into 2024 and beyond. That is a massive market for us that by design is a systematic market, which is why we launched a system platform and not a chip or not just the software this week because none of the automakers of the Tier 1s want you to deliver a piece of the puzzle. They want to solve real problems all the way down to the end user. And so that will add another $4 billion to $5 billion to our TAM, or SAM I should say, beginning in about 3, 3.5 years. And that is a category that we're incredibly confident about. I think we hit the market just at the right time. The way we solved it, I like to say this around Qualcomm is we're really right to left thinkers. We don't think left to right. We have technology, how do we make it important to the world? We really think -- we sit down with automakers and say, "Okay, what is it we're trying to -- let's co-vision. What are we trying to solve? What are we really trying to do?" Well, bring us a supercomputer. That's not a problem. That's not a problem that you don't have a supercomputer. It's a problem that your consumers want -- we want to save some of those 1.2 million lives that get exhausted on the road every year. Maybe that's a problem we can solve. And so that's what we've done with companies like GM that we announced this week, Jaguar Land Rover. Many of the ones, PSA, many of the ones that we've announced over the previous years were really right to left. We work with them right to left. And frankly, half the time when we're working together, I don't know who works at which company, we're just trying to solve for, how do we -- 2 things, how do we bring the digital life into the car smartly in an automotive way, non-distracting? And then how can we really get safety to the next level? Not all the glitz and glitter, how do we really bring safety solutions in the car in a meaningful way at budget in a thermal way that makes sense in your trunk that looks more like half a box of golf balls, not your kids' video game server? And if you go to the Qualcomm booth over in Las Vegas Convention Center, you'll see our Level 1 to Level 5 scalable solution. It's this big. It's about that tall. It's about that big. You can hold it in one hand. And that's what we've been grinding away on for the last 5 years. And we do the hardware and then we also do the algorithms. So that's where the opportunity is coming from. And I'll be honest with you, every -- a quarter does not pass by where I don't get another business plan on my desk saying, hey, these other assets, GNSS, we can deploy GNSS, which we did. We can deploy RF front end from that side of the house. And so as the general manager, leader of the automotive business, frankly, I feel a little bit -- it's an embarrassment of riches because I receive a lot of the technology that we develop on the other side of the house, which, frankly, because we're a mobile company is evergreen. Every year, our GPU, our CPU, our AI, it's all refreshed with the latest consumer needs. And I can take that and run with that at speed. And I think that's a big key to our success, our ability to move so quickly.

No, that's helpful. Let me check if anybody in the audience has any questions at this point. Maybe then, Patrick, if I continue kind of with that theme overall. One of the things that you kind of did not bring up but kind of was mentioned yesterday and kind of what you're implying in the connected and telematics growth is V2X. And kind of even -- I remember like coming here 3, 4 years ago and V2X used to be a big demonstration even at that point around kind of the dry run drives in Vegas. So what do you need to see in terms of changing either in the regulatory landscape? Or even if we -- like investments from the kind of local authorities in the equipment to get V2X to a point where it starts to inflect? And how do you think about kind of that trajectory from here on because V2X has been talked about for a while now, but we haven't really seen those applications or those use cases kind of as V2X usage on automotive vehicle yet as much?

Well, I'm thankful that you asked that question because if you asked me, of all the things that Qualcomm is doing in automotive, $600 million business this past year, multibillion-dollar business in our very near future, the thing that I'm most proud of is the V2X technology that we're developing, which is an offshoot of our telematics business. And the reason is this. 10 years from now, your kids or the people you're talking to will get a good belly laugh that we used to drive around like cowboys, independently just throwing our neck, looking around to see who's around your car. Your car will know whether it's safe to turn, whether it's safe to go through an intersection, even when there's a green light. C-V2X is a technology that gets an electronic picture of everything that's around the car and really stops you from doing ridiculously silly and stupid things. And so it's such a great technology that I think -- we're so passionate about making this technology commercializable as soon as possible because honestly, it's -- I think it's going to be one of the biggest things to saving over 1 million lives over the coming few decades. And so it does connect the car to the world around it. It connects cars to where cars can talk to each other, where cars can talk to infrastructure, where cars can talk to the cloud, not only for the convenience of maps and things like this, but also, if you're not using your blinker and you decided to go left and I decided to go right, which happens all the time, our cars will know and they'll stop each other from doing silly things. The biggest challenge for us initially was getting all the automakers on the same page, the industry, telecom and automaker all on the same page. That's been done, I think. There are a few that I think are holdouts, but not many. Most -- even the stalwarts that initially started with the competing DSRC technology, the 802.11p technology, most of them now realize that C-V2X is just superior technology in every dimension, performance, reliability, road map, you name it. That was the first hurdle, to get everybody on the same page. I feel like we're 99% of the way there, really across the globe. The next hurdle, and we're still clearing that, is really spectrum, allocating the spectrum to where -- there's clean spectrum for the safety use where cars can talk to each other in a relatively quiet environment. And so there is a hope of a global harmonized chunk of spectrum at 5.9 gigahertz. China jumped the gun early and I think in a very good way, allocated 20 megahertz of spectrum some time ago. And they're running -- they're going to deploy C-V2X with Qualcomm's help and cooperation this year. And so they'll start saving lives earlier than the rest of us. North America, the FCC just now proposed setting 20 megahertz aside dedicated for C-V2X, which is a huge win. So happy, that was a great day for all of us across the industry. And I think in Europe, there was quite a bit of debate, but most recently, they decided to be a little technology-neutral, putting down a delegated act that was just focused only on DSRC. So getting the spectrum allocated was a big thing. I think like most of that is -- we're getting pat -- through most of that this year. And then I think the next big hurdle and a point of one of our announcements this week was it's not just having the onboard units. You also have to have the roadside units. You have to have the infrastructure able to talk the same language. That roadside, those roadside units with cameras and other sensors are going to know, there's road work going on. You don't even have to drive over and have another car tell you. It's going on. We can just broadcast that right away. And so the roads will be talking to your car. Other cars will be talking to your car thanks to cellular V2X. This isn't just a Qualcomm thing. This is an industry thing. And frankly, this isn't something we try to do alone. It's through -- we work with the telecom providers. We work with the automakers across -- and Tier 1s across the globe. We work with DOT, FCC, governments across the globe. And so I believe that this is one of the most important things that Qualcomm Automotive will ever do in its history. And so we're very confident that this is going to be pushed through in its entirety over the coming 1 year, 1.5 years.

Any questions from the audience? Patrick, then maybe if I can kind of finish off with a couple of questions. One, if I kind of take the -- play the devil's advocate here on the Snapdragon Ride platform, probably people who've been coming to see us for a while would argue that most automakers today have been using NVIDIA platforms for their autonomous vehicle tests and NVIDIA has launched their platform quite a few years back. So as they now look at Qualcomm as an entrant, given your announcement yesterday, maybe like what is the differentiation that you can offer to some of the more entrenched players like NVIDIA in this space?

It's a good question. I appreciate that frankness. So in my gratuitous free time, I'm a baseball coach. And so typically, I'll tell the boys, this is a -- is this a drill that requires quality or speed? Which one are you going to focus on for this drill? And this is not one you want to focus on speed. First to market in a market that's going to take a dozen years to evolve is not an attribute, I think, of a successful solution. So this is something, quite frankly, as I mentioned, we're 5 years in the development here, maybe more on AI. And it's not something we felt compelled to come out with something -- we can throw horsepower. Qualcomm is one of the best in the world at throwing horsepower at things. If we want to throw out a 25-core engine, we can do it. But this is one where we felt really we want to work with the automakers and co-define value that was meaningful and actually, more importantly, practical, something that can actually be deployed. And so a couple of areas of differentiation. Firstly, is we're a full solution. We're an open platform though. We offer -- we have, from Level 1, Level 2, Level 3, all the way through Level 5 autonomy. We already have the road map that will deliver all levels of scalability. Now if GM wants to move at one pace and the other automakers want to move at a different pace, so be it. Our platform is modular and scalable on the hardware and the software level that can allow them to move at a pace that makes sense for them and their customers. It will also make sense on their low-end line to do Level 1 and on their high-end line, do Level 3, Level 4 or Level 5. And so one of the things the automakers really want is scalability. They don't want to do a low-end solution with this vendor and a mid-range solution with that vendor. Frankly, it's just too expensive. They can't afford to do it. So one of the primary mandates that they gave to me personally is I have to -- I want one big development where I can decide at what pace makes sense, hardware, software algorithms, the whole bit. And so you'll notice that our announcement is not hardware-centric. It's not algorithm or software-centric. It's the entire solution. If an automaker comes to me and says, listen, I don't want to use your algorithms, I want to -- I'll use these other algorithms, we will port those algorithms to the platform. If they say we want to use some of your algorithms, but not the others, that's fine, too. And so we're a completely open platform, which solved one of the competitive issue. One of our major competitors is a closed platform. That didn't go over very well. Automakers have their own idea about the experience they want to deliver. And what they don't want is to have a closed platform that gets forced on them. That didn't go over very well. Another area that I think is foundationally important and oddly overlooked was, again, everything is going to be EV moving forward, particularly if you look at the time frames that we're talking about. You can't have a supercomputer in the trunk for space reasons, for thermal reasons. Liquid-cooled 700-watt supercomputer is a non sequitur. It's just a nonstarter, period. And so that's technology coming out and trying to find a problem. The problem is that you have an EV out there where people are already range-sensitive and range-anxious. So you're going to have to deliver something that has a reasonable power budget, can be air-cooled or passive-cooled, and that has to fit into an EV and that's the needle that we had to thread before we made this announcement, so that's what we've done. Very power-efficient, very thermally efficient, systematically cost-efficient, and probably the most important is adaptable and scalable. If we have automakers that say I want to just start with Level 1, we'll see if my customers can get comfortable with that, we're done, let's go. If they say let's -- we want to do a Level 1, Level 2 combo, let's do a Level 3 and then we want to start to work on a Level 5 high-end solution that we can work with Google or Waymo, great, that would be great. And so that's really the difference between Qualcomm because I think -- I honestly think that our platform is the only practical, realizable, commercializable platform out there at this time. Everybody's going to adapt. Everybody is learning. We've been -- it's odd that several people tell me this was a surprise to them because we've been driving around in huge black cards that say Qualcomm all over them with sensors all over them around San Diego for -- since 2017, but apparently undetected even though we also had a license for that. But that's where I think we're different. And I'll tell you, there's 2 things that Qualcomm, I think, does differently. The first one is the assets that can come to me from the other parts of the house and the leverage I can get off these assets allows me to both be accretive as a business unit, but also move at the speed of innovation of the new automotive industry. The other one is this right to left thinking. We really do everything in partnerships with automotive. We never launch anything or announce anything until we have multiple customers already saying that's what we're going to do together. And so -- which is true for every one of the announcements that we made this week. And so I think this trusting in the partnerships and this cooperation with the leading automakers, I think, is really allowing us to deliver stuff that makes sense, not technology for technology's sake, but stuff that solves real problems.

Great. That's all we have time for Patrick, but thanks for coming to our conference. Thanks for participating. Thank you.

Thank you. I enjoyed it. Thank you.