Mark Odendahl
Welcome back everyone and welcome to the Industries in Motion podcast from RBC Capital Markets. A reminder this podcast is where we explore what’s new and what’s next in today’s fast moving markets and industries to help you stay ahead of the curve. Please listen to the end of this podcast for important disclaimers. My name is Mark Odendahl, and I am Head of US Capital Markets Research. I’m very help to have Tom Narayan on the podcast today. Tom is RBC’s Global Autos Analyst and he recently published a report or a primer on robotaxis and
and how robotaxis are going to develop in the auto industry and in global transportation. Tom, welcome to the podcast.
Before we start, I would like everybody to know a little bit about Tom. Tom started at RBC as a research associate here in the US, then moved on for a lead analyst role in Europe a few years ago where he started covering the European automakers and auto parts makers. And just recently we presented Tom with the opportunity to move back to the US and cover the auto industry globally. We think this is unique among Wall Street analysts to be able to cover the significant automakers in Europe, combined with the automakers in the US and educate investors around the disruption that the EV development is giving the global auto industry, Tom, let's get started on all the Q&A today about robotaxis.
Tom Narayen
Sure. And, autonomy is a very kind of nebulous kind of term we've actually had autonomy for, for several years, most of us have it in our cars today. It's things like Lane Assist, emergency braking, cruise control. The industry, when we look at it, we really break it up into kind of five levels but really the first three levels these are where you need a human being that can, when needed, actually drive the car. And then you have two levels at the top level four, and five. These are where there is no human being needed to drive the car, any person in the vehicle is simply a passenger. Today in the developed world, we're probably at level two going to three. We're not at the level or the stage where we don't need a driver.
Mark Odendahl
So maybe talk to me about some specific OEMs, and where some of those OEMs are. And relate that to some of the bigger picture.
Tom Narayen
It is interesting when you look at the OEMs, the premium OEMs are probably farther along as you would expect with autonomy. Notably, companies like Mercedes, which became the first ever OEM earlier in this year to get level three regulatory approval in hazard in Germany, California, Nevada. And that means you can drive the car in certain conditions only on highways below 60 kilometers an hour. But you don't need to have your eyes on the road or your hands on the wheel in those certain conditions. You have BMW, which just recently got approval in Germany to have level two plus, which means you do have to have your eyes on the road. But you don't have to have your hands on the steering wheel up to 130 kilometres an hour. And of course, you have Tesla, which has its FSD. It calls it full self-driving. It’s FSD system, which is a level two plus system, it's allowed in the United States, you have to have your eyes on the road, you don't need to have your hands on the steering wheel. For the most part, though, most OEMs, especially in the mass market, are at base level two systems, some of the stuff we talked about earlier, emergency braking, some park…medium parking assist, a cruise control, some lean assist, but it's really these premium OEMs and Tesla that are leading the way.
Mark Odendahl
So the viability of autonomous vehicles has been hotly debated recently. Could you sum up some of those debates and challenges that we've had in the near term? And then kind of marry that with some of the long-term views?
Tom Narayen
With autonomous vehicles, the main hurdles are two things. One is the tech. And the second is the regulatory hurdles. So we think that you'll have the tech actually solved first, and then the regulatory hurdles, usually regulators want to see a system that is close to 100% success rate. The problem that we're having with level four and level five autonomy so far, is that it doesn't work very well. in urban environments, there's a lot of random stuff that happens, right? Pedestrians, you know, animals, bicycles, etc… other humans. But we do find, and this is something that's very encouraging, is that on the highways, we're already pretty close to being there. In urban air environments, we're pretty far away. We think, though, by the end of the decade, we'll get technologically there for level four in many large cities. And then regulatory, we think we'll need another few more years, that'll probably happen we think by 2035.
Mark Odendahl
For a regulator, what are they going to be happy about and excited about? And what are their concerns going to be?
Tom Narayen
It really depends for a regulator where you're looking at. So in the US, it's really a state by state decision, you know, California or Nevada or Florida, or even Texas may be more aggressive to approving it than, let's say other states. And in Europe, it's probably going to take longer than any other place. And then China, we see regulators are approving it quicker than any other place. So it also has to do with an attitude of utility, how the society views, the utility, unfortunately, some there will be deaths. And societies that we find in China, they're more willing to accept this is better for the society as a whole. Some states in the US and then, you know, later on, you'll see it happen in Europe; but we think what will drive the regulatory argument is how many lives it'll save, you know, every year, 2 million people die in car crashes. 100 million people are hospitalized, these vehicles will save up space in cities, a third of cities are used for parking and looking for parking. The average commute speed is 10 miles an hour. So it'll save up hundreds of millions of hours of productivity for people. So we think policymakers, there's inertia for them to support it. But they will need enough of a technological argument to showcase that, hey, you know, 94% of accidents are human error with, you know, these vehicles, it drops to like, two or 3%. So it needs to be a sizable number. But there's enough inertia for this to happen. And if it happens quicker, I do see politicians acting on it quicker.
Mark Odendahl
Let’s continue to walk out the end of the curve here and talk about robotaxis. How far out is that inflection point? And what role do you see robotaxis serving?
Tom Narayen
It is it is a long curve, admittedly. So 2030 is when I think you'll start seeing the tech being, you readily available, where, the accident rate really is low enough that OEMs are willing to take on liability, because it's so low. And then 2035, I think you get the regulatory approval. But it also becomes ubiquitous when there's an economic argument for it. Currently, we believe that the cost to make a robotaxi is well over $100,000. You know, the cost to make a mass market car is something like, $20,000, right. So there's a huge spread in the cost to make the vehicle because of all the sensors and the tech that goes into it, such that the cost per mile cost for a consumer or for an operator of a robotaxi it's just too high right now. Companies that have services for robotaxis are losing money. But we think by the end of the decade, the cost to make that robotaxi could drop dramatically and effectively cost the same to make, maybe slightly more, but to make a regular passenger car. And on that basis, our math actually suggests that, because it's increased usage and the utilization of robotaxis, it increases so much. Actually, you know, a passenger car right now is only used 5% of the time, most of the time, it's just sitting in, you know, parked. So the robotaxi utilization, and the fact that it costs the same to make the car drops to like 30 cents a mile. Whereas the cost of owning a private car is 50 cents a mile. So it's actually cheaper to own and operate a robotaxi than to own a car. And we think once that economic argument is made, that's when companies like, Tesla, or Waymo, or other companies, fleet operators will come up and try to make this business work. So it really it also has to be an economic argument, in addition to a regulatory and tech one.
Mark Odendahl
You made the argument for the economics of a robotaxi. But what about for the OEMs and margins and their ability to make money? You know, it sounds like it's not an economic good decision for all of us to have cars. So that can't be good for OEMs longer term.
Tom Narayen
I think what happens is, there will be winners, there will be losers. So those folks who really go into the robotaxi business head first they maybe try to own their own fleets or participate maybe making the actual robotaxi, maybe they don't operate the fleet, but participate in this business could be handsomely rewarded, right? Because we've seen the profitability, it could be massive, right? Today, when you take an Uber, you pay spend like $1.90 per mile. A private car, it costs you about 50 cents to own, but people are still willing to pay so much more because of the convenience. So imagine the cost of the robotaxi’s 30 cents, and you're able to charge, 90 cents, the margins are massive. So it can be hugely profitable. So the other thing to note is everyone robotaxis replaces five private cars. So if you're an OEM, you sell so many less cars, right? So you would think that would be bad. And just on that basis, it would be bad, you're selling far less cars. But the flip side is, because it's so much more profitable, right? The auto business you're making like what mid-single high single digit profit margins, operating a robotaxi, you could be making 50%, 60% profit margins. So you may be selling less cars. But you could be making way more profit from running and operating this fleets. The only issue is, you know, how many OEMs do we have? 30/40 car makers? Do we need that many companies making robotaxis? Like when's the last time you cared what Uber picked you up? You don't care, right. So the value of the of the brand, maybe gets more commoditized. So a lot of smaller OEMs may go away. Maybe there's a lot of consolidation. But we do know for a fact it will mean less cars on the road, less cars made. And these OEMs who are best positioned to capitalize on this theme could make, money hand over fist. But there also will be some companies that simply get gobbled up in this consolidation fest.
Mark Odendahl
All right, let's stay there and dig a little deeper. Could you educate us a little bit on the strategy that Tesla has to capitalize on the growth in this space?
Tom Narayen
So in recent months, what we've noticed from Tesla is a willingness to partner up and sell services to other OEMs. Typically, that doesn't happen. Usually, this is why suppliers exist. And we saw this happen with the charging infrastructure deals that Tesla made with Ford, GM, Rivian. And some others. A lot of automakers are selecting Tesla's charging infrastructure in the US specifically and we expect to happen in Europe as well. Tesla is doing this on one part in order for it to sell its FSD software to other OEMs; so it's using the car as a Trojan horse of sorts to sell software. And the way that Tesla looks at the car is they already have the best cost economics on full electric cars, because they started before everybody else. Their costs are around $100 per kilowatt hour, whereas others are $150 per kilowatt hour costs way more. And if you look at their margins, they're 13 to 14% EBIT margins, whereas, you know, Ford, for example, their EV EBIT margins are like negative 40%, simply because they started later. So they have a better cost structure. And therefore, what happens is they can afford to cut pricing, which is what they've done to gain market share. So their argument is, they want to sell as many cars as possible, gain market share and put FSD in those Tesla's, which they believe they can make more profit off them, right, because it's a software margin. Currently, FSD, they charge about $200 a month. So their plan is to cut the pricing and increase the take rate of that subscription, which maybe it means that yeah you get less money for the FSD per car, but you get more cars taking FSD. And the other thing it does is, and this is where the charging infrastructure deal happens, it means they can sell FSD to other car makers. I don't know if they're gonna win all of these wins, other companies are also doing this, you have Qualcomm and Nvidia, Mobileye, they're also in this space, but certainly it gives them an ability to penetrate this market. So that's really how they're approaching it early on and then longer term they want to be heavily in the in the robotaxi space themselves. They're big believers in this, and they believe they can not only make their own robotaxi fleets, but also license robotaxi software for carriers down the road.
Mark Odendahl
So Tesla is clearly ahead here and forward looking. How do you think about market penetration in this sector going forward?
Tom Narayen
Tesla definitely, I think has the leadership position on FSD. And has the most miles on the road, it doesn't mean that others are not progressing. But you know, my analysis only has them capturing 25% FSD, in robotaxi penetration in the US only seven to 8% penetration of their own vehicles in China and Europe. And on the licensing of the software I have them maxing out at 20% market share in the US, China and in Europe. So that does give room for other folks that we spoke about Nvidia, Qualcomm, Mobileye, Google, Waymo, etcetera. So I don't think that they necessarily dominate the market. There are others that are, making strides, but I think they will be a substantive player. And the other thing I would add is, since things like robotaxis are so far away, to capitalize on this, you need to basically be able to take losses, spend a lot of capital for many years, and capitalize on this whenever it comes. And so a company like Tesla, which has this giant market capitalization, Google Waymo, which has a giant, balance sheet market cap, maybe Apple, these are the companies that can afford to last that long to see that light at the end of the tunnel. We all saw what happened with Ford and Volkswagen with their Argo joint venture, they couldn't fund it so they closed it last year. Another factor is those folks who are well capitalized enough to survive, they may be the one that really sees the fruit of that investment at the end of the day.
Mark Odendahl
Well, Tom, this has been a fascinating conversation. And I think we could talk forever on this. And the whole opportunity is open ended when you look at some of the things that autonomous vehicle could do for our society and do for the economy. So really appreciate your time today. And thank you very much for all the work that you did, putting this robotaxi primer out.
Mark Odendahl
What else lies ahead in today's ever evolving markets and industries. We'll be keeping track right here on Industries in Motion, Until then, thank you for joining us today on this episode recorded June 30, 2023. Make sure you subscribe to Industries in Motion wherever you listen to your podcasts. If you'd like to continue this conversation, or you're interested in more information, please contact your RBC representative directly or visit our website at www.rbccm.com/industriesinmotion for further insights. I really enjoyed talking about the auto industry with Tom today. And I look forward to seeing you at our next podcast. Thank you.