General Motors Company (GM) Earnings Call Transcript & Summary

We're thrilled today to have Gil West, the COO of Cruise with us. Prior to joining Cruise, he was also the COO of Delta Airlines, and he just flew in today.

Yes, just flew in today.

Thanks for joining us.

Yes. Great to be here.

All right. Let me do the disclosures. You can get your [indiscernible]. We'll hand all that. Please note that all important disclosures, including personal holdings disclosures and the Morgan Stanley disclosures appear on the Morgan Stanley public website at www.morganstanley.com/researchdisclosures. The disclosures are also at the registration desk. Some of the statements made today by GM may be considered forward-looking. These statements involve a number of risks and uncertainties that could cause actual results to differ materially. Any forward-looking statements made today by the company are based on assumptions as of today, and GM undertakes no obligation to update them. Please refer to GM's Form 10-K for a discussion of the risk factors that may impact actual results. So maybe we should start a little bit with your background. I would first love to hear what attracted you most to the Cruise opportunity. And when you sort of -- when you do a lot of meetings and you sit down, you talk with investors, what is sort of the biggest part that is still most misunderstood?

Yes. Well, let me just say I'm a transportation geek at heart. So it didn't take much long to us to join the Cruise team given we're creating a whole new mode of transportation. So that's the big draw for me. I've been here for a little over 2 years now as COO. And I'd just say really just living the dream in terms of probably the most advanced application of AI that the world's ever seen with self-driving cars. So it's an exciting space to be in. I think before I joined Cruise, as you mentioned, I was COO at Delta Airlines. I was there over a dozen years. And that was -- I joined about just after bankruptcy there. So team sport that we took the airline from bankruptcy to kind of the pinnacle of the industry. So reliability, customer experience, most profitable, most value, so I think some of those lessons transfer over to Cruise, not all, but I mean, aircraft are not autonomous vehicles, but there are some things that overlapped that in parallel. So safety, of course, a big part of the equation. I think the airlines are probably the pinnacle example of safety management, risk management. So we've adopted best practices with safety against everything we're doing at Cruise. Reliability of course, customer experience, really cost management efficiencies, leveraging your fleet, high-asset fleet to drive utilization, those type of things. So there's clearly some parallels with what we're trying to do at Cruise.

Transportation economics.

Yes.

Yes. I know my team saw a couple of them rolling around the street last night. Maybe I think it's been about 1.5 years in San Francisco now. Can you just sort of talk to us about some of the early learnings over that last 1.5 years in San Francisco, both on the positive side as well as sort of hurdles you said? Okay, these are challenges. We really have to overcome as we think about scaling to new markets like Austin or Phoenix, et cetera?

Sure. Yes. So we -- just to give a little bit of context around it. So we launched full driverless service a little over a year ago. Since then, we've accumulated over 1 million miles of driverless miles. This is no safety driver but driverless miles. We've added the public in the cars as well, and of course, continue to expand. So there's a lot of lessons and learnings flow from that. I would probably bucket it into 3 different buckets. First is the customer experience as we brought the public in. So just refining that based on feedback, as you would expect, we've got good data sources from our customers. So generally, it's continued the app development. This is an app that you would use to hail a vehicle just like you would an Uber as an example. So that app development through the eyes of the customer, that development has continued, giving the customer more control over the experience is another big piece of the equation. So especially the kind of in-cabin experience in terms of being able to just end the ride, if you get to -- where you think it's a nice destination you want to stop without having to talk to somebody, being able to control the temperature, play your music, variety of different things. So that piece has been, I think, instrumental in the development. And then reliability is another bucket I would add. So again, it's a lot different. The signal you get is a lot different moving from 2 autonomous vehicles to a couple of hundred, right? So you start to see trends data starts to normalize and you understand where the opportunities are, so that's been an area too of learnings. And then really, I'd probably say scaling is another big area because I think just to get a better appreciation as we've begun to scale in terms of kind of the constraints to scale, the interdependencies in order to be able to basically work those off so that we can unlock our ability to rapidly scale. So we're trying to institutionalize a lot of that as we move forward.

That point around scale and sort of constraints to scale, it's sort of -- you mentioned earlier how there's such a complicated engineering problem. Are there any learnings you've had where you can sort of increase the pace at which you expand the service radius? Like, how should we think about it? How fast -- how long does it take before you can fully throughout the San Francisco Metropolitan area? And then learnings, which you can go to Austin, Texas and say, we will start with a much larger service radius than we did previously?

Yes. So again, a little bit of context. So we started in San Francisco to develop the tech here because I think we viewed it as the most complex urban environment to kind of flush the technology out. You hear about edge cases, et cetera. You guys walk around the street here, you see plenty of the edge cases. So it's a good environment to develop in. And so we've learned quickly. We've expanded quickly both what we would call the operating domain throughout the city. Time of day is another variable. So as we went to daytime operations, fully driverless, there's a number of variables that we have and levers that we have as we expand number of AVs in a given point in time. And what I would say is everything we do is gated through safety ultimately. And as we -- and then we learn based on safety, reliability, all the signal that comes in. And it's really what I would describe as a continuous learning machine. So it continues to mine the data, train the neural systems that we use and basically improve the stack so that it's better and better and better as time goes on. So there's a continual flywheel model, and then we can take that to other markets as well. So track the stack definitely is extendable into other markets.

Okay. Earlier on, when you mentioned 3 of the important points around customer experience, more in-app control and then reliability, one of the questions I always had about how quickly could you scale. How do you think about the idea of partnering with like a rideshare network that already has human drivers to fill in for a lot of the use cases and then you could offer users and say, if you'd like to have an autonomous ride, that's at a fraction of the price. Within this radius, you can. But then we also have millions of people who can drive you anywhere. And over time, you can sort of expand the autonomous radius. Is partnership the way you think about it? Or is it more we want to stay top of funnel?

Well, I think we're open to partnerships. Let me say that partnerships are kind of fundamental in the way we're approaching the business in different areas than the example you gave, but I wouldn't necessarily rule it out. But just first on partnerships, our ability to scale is really enabled through partnerships. So we've got -- as we think about it and as we go to markets and as we scale, we think we can do this with minimal incremental spend because of the partnership. So infrastructure, partnerships, charging real estate, data offload, et cetera, as well as service partnerships, so maintenance, et cetera, and then, of course, we've got a big partnership with General Motors around the production of purpose-built vehicles. So all those partnerships are kind of fundamental in the way we operate. And I think we do have a partnership model in our DNA. As it comes to go-to market, there, I think what we have and what we'll deploy shortly with our first purpose-built vehicle, the Origin, is, I think, a much better product at a much lower cost. So I think we asked ourselves, with that model, does it make sense to partner or not, especially as we move into this because our experience to date, the customers just love our R&D vehicles. It's better experience. It's consistent. And then as we move into the Origin, a few of us have had the pleasure of riding in it. It's game-changing. I truly believe we'll have a -- you fast forward a year from now, we'll have a cult following around the vehicle. So -- and then that's the form factor that we start to drive cost out of the business with. So -- we look at those factors and we go, why wouldn't we go to market ourselves. So that's been our approach on the go-to-market.

Okay. Let's talk about the technology a little bit. And I think we can maybe go into sources of scale over the long term as well. There's a few different philosophies about the way to think about the technology behind autonomous, whether it's RADAR, LIDAR, mapping, no mapping, et cetera. Just maybe to level set everyone in the room, just remind us of sort of where Cruise sits on each of those key areas, why you chose those and then if you want to go into sort of, as you scale, what are the sources of leverage in the model across those decisions?

Yes. Good question. So I think as we think about autonomy first, just philosophically, the way we're approaching this is we're looking to develop the driverless performance as quick as we can. So that's kind of where we've emphasized. So we'll look for basically any sensor feature that can accelerate that. So we've deployed maps, cameras, RADAR, LIDAR, there's no one sensor does it all. So it's a combination of the suite to get the 360 view of the world. But we're always looking and iterating, right? And the technology keeps developing. So with that in mind, what we've tried to do is accelerate the performance of the self-driving features in the EV stack. And then as we've been able to unlock the ability of the cars to safely drive themselves in this environment, then now the focus continues to be how then do we drive cost out of the business. So make the cars drive themselves, look at the tech, advance the tech, then how do we drive cost out of the business. So as we've done that, of course, we partnered with General Motors around the Origin vehicle, which is our purpose-built vehicle, autonomous vehicle, no conventional driver controls at all, that's truly remarkable. So that is an unlock for us then to drive the cost out of the vehicle. So as you play that forward, we'll continue to look at hardware, software, both in terms of component cost as well as the quantity of components that are on the vehicle and continue to drive cost out as we move forward. And then we're also fortunate we've got, I think, one of, if not the only now, Founder/CEO still at a serious EV company. So Kyle, our CEO, he's got a lot of super powers. But one big one is that he really sees where the tech, where the puck is moving in terms of the tech. So we're able to really play that out forward and then help understand what tech should we be pursuing, how should we apply it? Because everything is about application and what we're doing. It's not just building tech for the tech stake. It's how do we apply it and really have a disruptive business in the process.

And you sort of talked to the hockey puck analogy and sort of how slippery is the ice, how fast the puck will move in the next few years. I guess maybe let's take a look back a little bit. I think that autonomous driving generally probably hasn't progressed as fast as a lot of us thought in the last 5 years for a variety of reasons. What in your mind are sort of the 1 or 2 key reasons why things have gone slower than some would have expected in the past? And then going forward, what are sort of the key macro gating factors are you going to determine how fast we get to autonomy?

Yes, great question. So look, again, I've been in the business a little over 2 years. But from my perspective, it's damn hard to make the cars drive themselves, right? And we've been able to accomplish that, right? I mean you go for a ride tonight and see you if you haven't already. So the tech is here, it works. I think -- so as I look at it then, we're -- we've got an execution DNA at Cruise. So part of it -- I think there's a mix, a special mix of things that lead to success. You got to have the tech, right? You got to have the talent. You got to have capital and partnerships, which you've got to be able to execute. So I think a combination of all those things are pretty powerful. And if I look at where we're at now, tech works, that's been, in my opinion, the biggest variance in the whole thing. Last year was a big year for us, right? We launched full driverless first time. We were the first company to ever launch commercial paid ride hail in a major U.S. city, million miles now complete driverless. So really for us, it's -- there's a lot of barriers to entry behind us now. So a lot of that variance is in the rearview mirror, so to speak. But as we look forward, I really think this year is really a pivotal year for us. And I think we basically work through most of the remaining barriers of entry this year. And the way I would describe them first is regulatory. So we should finish out all the federal as well as the California regulatory requirements. So that's a big check. And then scale is another big one. Scaling is hard, it really is, and a lot of lessons. But we're -- we've, I think, developed a playbook that we've demonstrated as we move from San Francisco to Austin and Phoenix in less than 90 days from scratch and stood those up. So we know the tech can move into more markets, and we've also developed a playbook on how we do that through partnerships to scale. And then really, the next big -- and of course, all that means more rides, more deliveries, we're in not just the ride hail space but the delivery space, and then we start to -- really start to increase our revenue flywheel. But the other big unlock for us this year is landing the purpose-built vehicle, the Origin. I would say probably, right after making the cars drive themselves, the next toughest thing is to build a purpose-built vehicle. So you're -- you've kind of got a partnership between a tech company and an OEM, both a lot of real strengths. But the way I've looked at it is they run at different gearing ratios, right? So that -- there's kind of a -- there's some magic to put those together. And I really credit GM team, Kyle, and the Cruise team, to be able to make that work. It wouldn't happen without Mary Barra and the GM leadership's team top-down support to make things work. So it's really hard to design and develop a purpose-built vehicle. But then we've not only done that, but we've got the supply chain. There's been investments at Factory ZERO at -- in Michigan to build the vehicles, it's tooled up, it's ready for full-scale production, right? So we're in the final stages of certification. So regulatory, scale, purpose-built vehicle, and then all that effectively is a combination to unlock most of the barriers to entry for us as we go forward. And then fast forward through the years, and it's more of the same, and we just continue to add zeroes. And we scale from hundreds of vehicles to thousands of vehicles this year. So I think this is a real pivotal year for us that I think will really transform not just Cruise, but the whole perception of autonomous vehicles.

As you've rolled out the full driverless in San Francisco, I remember over the years, there was a discussion about sort of the psychological barriers for riders to feel comfortable taking a driverless car on the highway, in heavy traffic, going 80 miles an hour, et cetera. What can you share with us about sort of user behavior or cohort behavior as they've gotten more comfortable with the idea, "there's no one behind the wheel."

Yes, good question. So how many of you have actually taken a ride? All right. So see if this fits with your experience. Virtually, everybody I've talked to that's been in a car, which is now a growing number of people, after about 15 minutes, you're almost back on your cell phone again. It normalizes quickly, which is exactly what we want. I think there's -- probably the first time you take a trip, there's certainly an anxiety. I equated it to the first time I ever flew. It was like kind of an out-of-body experience because suddenly, this car when nobody in it pulls up, you get inside it with your phone, and it drives you away. It's kind of like holy s***, what is this about? And then -- but honestly, quickly, it just normalizes. It's like, okay, this thing's got it. It's safe. But it's able to navigate situations that human drivers would really struggle with. So I think all of that just -- I think it's something that really just normalizes in a hurry. And then once you do -- certainly, once you do 1 or 2 rides, it becomes normal experience. The nice part of that, I keep saying it's a better experience, but I think it's -- and the -- just with our Chevrolet Bolts that are here is R&D vehicles, certainly with the Origin, it's at a completely different level. But just even with our R&D vehicles, it's consistent. You don't have another human in the car. You don't have the bizarre smells that you may encounter in some of these rides. So it is -- I mean, it's definitely a different experience. And I think that part, coupled with the fact that you realize fairly quickly the car is safe and has control of its behavior.

Okay. Let me ask about the losses a little bit. I think Cruise lose about $2 billion this year, maybe half of which is cash. How should we sort of think about the loss levels and sort of the cash investment levels as you go forward and sort of roll out the Origin and I mean at what sort of scale to the unit numbers you were talking to earlier?

Yes. So I mean the way I would describe it, scale matters. Scale matters in our business, drives revenue and ultimately profitability. So the sooner we scale, the sooner we overcome our overhead cost and reach profitability. So -- and we're positioned to scale as the bottom line. So as we enter new markets and scale, when you think about cost again, I think I mentioned earlier, but not a lot of appreciable incremental cost for us to scale because we're doing it through partnerships, other folks' capital, but we're also driving disproportionately lower operating cost at the same time because we're bringing scale to the table as well. So as we enter new markets, we're able to do it efficiently. But then also the other big variable is vehicles and vehicle costs. So I mean, we've made -- we've already made the investment to develop the purpose-built vehicle. This isn't something we've got to do. We've done it. We tooled up for it, plant there, so we're ready for full out production. So scale matters. We're positioned to scale quickly. And then the economics are very strong as we start to reach scale. I think the other thing I would mention is we're in 2 good initial business segment, so ride hail as well as the delivery business. So I think if they're prime for autonomous application, right, you take the driver out of the equation, you basically are driving long-life vehicles. So we're able to amortize our life over 1 million miles, as an example. And then really, we're able to drive a much higher utilization rate than the current models, right? So I think the economics bode well. And then I think the tech as well that I talked about, I think through the partnership with GM, we're able to leverage that scale, continue to drive cost out of the business. And then the markets continue to open up as we do that as well. So the traditional ride hail market really starts to expand as we drive cost further out of the business so that it starts to grow to the point that it starts to actually take some personal car ownership into the model.

If anyone has questions for Gil, raise your hand and we have mic runners going around. There's one in the front row here.

Well, super interesting. It sounds like a lot of progress over the last 12 months when we were here last year. So where does it leave Uber? Like if you -- you can go straight into ride hail and delivery with higher utilization, purpose-built vehicles. I mean how do you see the existing infrastructure in that space dealing with this new competitive dynamic?

Yes. I think -- I mean I can't -- obviously, I can't speak for Uber, but just looking at it through our vantage point, it does seem like the right go-to-market strategy for us. If you're a cost leader with a better experience, makes sense to go to market that way. I think we'll -- we're entering the first year of our rapid scaling, but it's a very big market, right? It's a global market as well. So I think the first number of years, it's a high-class problem, but we'll be supply constrained even with the probably best partner has been producing scaled vehicles for over 100 years, right? We've got it geared up, but still it takes some time to get critical mass. So we're comfortable with our approach. From a product market standpoint, the customers love it. I mean we've got very high Net Promoter Scores and star ratings, so I think the people that have tried it. For us, so it's building it -- building the availability side of the equation now. It's the other part that we're very focused on. So we know got to have it when people need it around the clock in a geography that makes it a dependable mode of transportation. So we see the path there, and we're going for it.

Not to have you -- I don't want to have to talk to any specific competitors. But it's interesting because we've seen other competitors sort of deemphasize programs a bit more. When you think about forward with Argo or even Waymo has had some restructurings, and Alphabet has quite deep pockets. Just sort of maybe stepping back, what are sort of some of the key competitive differentiators for you as you were sort of evaluating? Going to Cruise versus any of these other players at that point, do you think have sort of enabled you to not go down the path that some of the other companies have?

Yes, great question. First of all, the partnership really important. I think it's -- I mean it's a key strategic advantage to have a partnership with GM and have the ability, again, to make the partnership work is really important. So I think it's a key unlock for us on profitability. Again, it's one thing to develop the tech. It's hard. It can become your whole focus. But to think commercially, how to apply it, is a completely different ball game, and then have the ability to do it is even harder. So the partnership is important for us. I think you're seeing just kind of the competitive landscape sort itself out naturally. I mean it's a separation of the winners and the losers in the marketplace, right? And I think folks with the right ingredients, the talent and the tech and the capital and the partnerships are winning and accelerating while others are dying, so I think it's just a cycle that fuels itself. So we're seeing that happen. But again, I'm here to tell you, and there's plenty of proof points, if you walk around the city, the tech is here. It's now. It's deployed. So now it's just a matter of scaling is a reality, continuing to drive the cost down to produce the profits that we're going to produce. So I think that's the way I see it is that. At Cruise, what -- the proof points for me that had me take the leap was the partnership and the execution DNA. It wasn't just about technology. It was about the application of technology, which I'm personally obsessed with. So I think it's the right combination. It's huge markets, the economics at scale or -- it's incredible business. So I think all the necessary ingredients are here and the ability to execute it exists. So just happy to be a part of it.

Anything you guys shared at all on the price per mile, you have to charge and severance is going to make a breakeven?

Well, I think probably the way I would describe it is we know that the markets open up really, not just here but everywhere, once you get into the $2, $1 a mile price point, right? And that's ultimately where we want to be. I mean, we want to be at that level. We've got a good line of sight to get there. The team -- again, the talent in the team's great at Cruise. There's a lot of experience, managing cost, managing both at the product level and the operating level as well as utilization of assets. So I'm convinced we've got the trajectory to get there. But then that starts to open big markets up. It's not just traditional ride hail. It starts to grow the market from both ends at that point.

Yes. A question over there. Yes.

Can you share with us kind of edge cases? And maybe out of your 1 million miles or total number of trips, how many have you had to have like human intervention or any type of kind of disruption to the trips and then how that is improving?

Yes. Great question. So we report all our safety statistics publicly. So I think if you looked at the last annual report, I think it was 700,000 driverless miles at the time. Of course, no fatalities or near misses. The -- so on a safety standpoint, we continue to improve and have been happy with the performance. The -- in terms of interventions, I'll just say that the EVs err on the side of safety, right? So when they encounter something and view it potentially as a safety event, then they'll err on the side of safety. So as an example, they'll pull over, put the flashers on, right? So then we're able to recover the vehicle either remotely or physically, in some cases, we may have to recover the vehicle. So that part of the equation continue -- we continue to learn that. But -- and for us, I think in the grand scheme of things, these are relatively easy problems to solve on the reliability front. Keep in mind, we've just got tremendous data coming in, so we're able to learn and continue to iterate and develop tools to manage this. But it's really about now industrializing the vehicles so that we're able to really build that level of reliability that's important to the business. So there's a lot of focus around that really.

Well, Gil, thank you so much. Can't wait for the next steps.

Yes. Exciting. Thank you for having me. Appreciate it.