Tesla's Cybercab Manufacturing Breakdown & Unboxed Process

[00:00] Let's break down how Tesla could potentially achieve five times the speed of manufacturing cars with their unboxed CyberCab products, which I've been really impressed by with its launch in Austin.

[00:15] And so what we've done is we've made a pros and cons sheet for Tesla stock via the CyberCab launch. But first, what I think is really interesting is just to preview is,

[00:28] and we'll really go through this in detail later, but there's an interview that we did about three years, two and a half years ago with ARK Invest Brett. And we talked about the future of the Tesla robo-taxi service

[00:44] and what it would likely look like. I think going back to dive into some of the expectations then is actually really important because there's a little warning inside of there for us. Now, this will make me less excited about the cyber type rollout.

[00:58] We'll talk about that in just a moment. But I want to now start with something that I'm really excited about. And it's this particular video here. This video is very exciting. And it reminds me of what I felt in 2022

[01:13] when I argued that Tesla should do everything in their power to copy and paste their manufacturing and just sell more Teslas. Elon later argued that he actually wasn't a fan of a copy and paste model,

[01:29] and what we noticed is that may have been because of declining sales in the S and the X, the Model 3 and Y were growing, but also a very kind of depressing launch of the Cybertruck,

[01:42] not selling a lot of vehicles. I still own my Cybertruck. I love it. I got the Cyber Beast. I honestly think it's one of my favorite cars. FSD is great. I feel like I never drive anymore. The steer-by-wire is fantastic.

[01:54] When you sit inside the Cybertruck, I hope I said Cybertruck, not Cybertruck. When you sit inside the Cybertruck, it's like a spaceship in there. It's so nice. I love it. It's a great vehicle. It's fun to drive.

[02:07] And it's a utility vehicle, too. But that said, the launch of its sales to other people, you know, maybe I'm a weirdo and willing to drive in the Cybertruck, but the launch of its sales to other vehicles just hasn't justified a copy-and-paste strategy

[02:19] where you want to mass-manufacture vehicles. Now, what makes me so enthusiastic, though, about the cyber cab is that it's possible that the cyber cab could go back to what I felt in 2022

[02:31] when I said, we can mass-manufacture this. Let's get to 10 million vehicle sales. Everybody should have these. Get the cost down to mass-manufacture. The cyber cab could be the tool that actually achieves that dream, not only through the unboxed manufacturing process, but through the actual utility of the vehicle.

[02:56] Now, the unboxed manufacturing process, and when I say my 2022 vision, just to be clear, the 2022 vision was actually a vision that I had after visiting Giga Texas about,

[03:08] wow, we could scale this. We can copy and paste this manufacturing process. The unboxed manufacturing process is actually part of Master Plan 3, the next phase.

[03:20] And the way it essentially works is rather than assembling a car, like this, where it goes down an assembly line and the whole thing rolls down this very long facility

[03:32] and slowly gets all of its pieces added to it. Rather than that, we're going to utilize this, what they call, unboxed facility. Where rather than moving the entire car down the line, we're just going to move the various different components down the line

[03:47] and then puzzle it all together like Legos. Elon today quotes that the unboxed manufacturing process could accelerate vehicle production by up to five times.

[03:59] That would mean taking an average Model Y production of one vehicle every 45 seconds off of the line down to somewhere around nine seconds. And some people are estimating currently, though it's hard to find official statistics on this,

[04:14] but some people are estimating they could potentially get down to five seconds per cybercap through this unbox strategy. And they call it 30% more space-time efficient, mostly because the assembly line is about 50% smaller in size.

[04:28] And of course, we've still got to prove this out. We'll talk about that in risks in just a moment. But the idea is that if you could manufacture these vehicles five times as fast, you could potentially produce five times as many of them with more operating efficiencies.

[04:42] And that's where you get to here. In 2023, they were calling for not only smaller footprints, but also potentially a 50% reduction in cost, which would translate to an increase in margins for Tesla, which is very exciting.

[04:55] Of course, we also had the announcement about no rare earths in the cyber cab. This was something that was part of Master Plan 2023. Now, it's been a long time coming. This was three years ago now, three and a half years ago,

[05:07] that we got this sort of Master Plan information. So where do we sit now? Well, now we sit in a place where people are starting to forecast massive enthusiasm around how quickly you can run one off the line, one off in every 10 seconds, theoretical long term, one in every five seconds.

[05:26] you've got people therefore now projecting that the cyber cab via an unboxed strategy could go and annually produce somewhere around two to three million units a year maybe even up to five million units a year at a five

[05:41] second cycle this obviously changes the dynamics for investing in Tesla stock because to me it always been about mass manufacturing the cars It always been about get these cars into as many hands as humanly possible

[05:54] be the next Toyota, where you could sell 20 million vehicles a year because you have the cheapest vehicle, the safest vehicle, the most reliable vehicle. Once you combine all those packages, you get really good sales.

[06:08] This is, I think, how Toyota has done so well. high quality, high safety, reasonable cost. The only way you can compete with Toyota is by competing on those three core components.

[06:22] And the CyberCab might be able to pull that off. Some risks associated with it. Let's talk about those, and then we'll talk about some pricing projections, as well as what ARK mentioned

[06:35] and what the warning was about the CyberCab. Going back to the interview that we did three years ago, it was really exciting. So let's look at the pros and cons. So here's the pros and cons table that I made. The first thing, 5 to 10 second cycle time.

[06:49] Five times faster production. The con is, can we actually scale to that speed? If the theoretical long term is 5x faster, potentially getting to 5 to 9, 5 to 10 seconds per vehicle,

[07:02] can we actually get there through the unbox process? That's untested. I actually think there's a good chance they can achieve this because even though that sounds really fast I don't think it's that unreasonable

[07:15] look at the math for a Model Y line and keep in mind that Tesla has the experience but look at the math for a Model Y line hopefully I did this right 365 days times 24 hours in a day

[07:27] times 60 minutes per hour times 60 seconds in an hour divided by 45 seconds per production gets me about 700,000 vehicles per year.

[07:40] But that sounds reasonable. You know, if I've got two lines, that's 1.4 million vehicles, right? So that actually sounds relatively accurate. So if I could 5x that, that would mean I could get to about 3.5 million vehicles produced of cybercats per year.

[07:55] I actually think the production is not going to be a problem. I think they could pull this off. Because even if they had a cycle time that was 20 seconds, so, you know, twice as fast, or frankly, even if the cycle times were the same,

[08:08] it doesn't matter. Then just have another line. As long as you're getting the margin of the vehicle, I'm not really worried about how fast we can produce them. Yes, if we can produce them faster, that's more upside to margin.

[08:20] Great, lower cost. Time is money. We all know that. Anybody who's played World of Warcraft knows... What is it? Is it the little goblin? Time is money, friend. Sorry, I get distracted. It's probably some ADHD problem.

[08:33] But anyway. the start of failing is the moment where people price in their most rapid enthusiasm for Tesla stocks because then the sky's the limit.

[08:46] Eventually we'll hit some kind of end of the S-curve where it's like, all right, we're going to settle around one every 20 seconds or one every 30 seconds, whatever. I don't think that part matters because I think they've got the experience and I don't doubt Tesla's ability to do that.

[08:59] What makes me more concerned is We're going to see margins start growing when these cybercaps really start getting to scale. Not immediately. They'll hurt margins at first.

[09:12] As they're scaling, you see the ramp of margins coming up. The impact of cybercaps will help lift margins across Tesla's fleet. And then the question will be, how long can we go with lifting those margins?

[09:24] That's going to be dependent on volumes growing, which we can only grow volumes when we finish edge mapping. So my thesis is edge mapping is probably one of the most critical factors for Tesla.

[09:40] We've talked about this in another video. But basically, the bottom line is how long is it going to take to edge map? So this is not like scanning the whole city and mapping it. It's going through all the edge case scenarios in that city.

[09:52] So it's sort of like mapping light, if you will. edge mapping took Austin, or it took about one year to edge map in Austin. Can we get that down to six months per city, three months per city, one month per city?

[10:05] I don't know. That is the critical thing that we have to solve. How fast can we edge? Okay, we want to get through the edging phase, and we want to get to completion.

[10:17] Once SSD gets to completion, and we can really sell these vehicles upscale, then the question of how many volumes we can sell long-term isn't going to come down to manufacturing or full self-driving.

[10:31] It's going to come down to profitability. And this is where there are a lot of unknown variables. So let's assume that in the long-term average,

[10:43] if you're selling 3.5 million of these vehicles a year, we end up compressing the value per ride to $1 per mile. The current Uber, you know, runs about $1 to $2 per mile.

[10:58] There are only 3,700 Waymos that operate primarily in dense cities so their cost per mile is much higher but I think that's because there's a limited quantity of them there's a novelty effect and they're in big cities.

[11:11] You know, if you're driving a mile in Santa Monica it could potentially take you an hour. It's not like driving a highway mile as in Texas, right? So if we don't introduce $3.5 million, let's say, RoboCaps, we're going to compress this to a dollar or less.

[11:28] That where we run into some potential problems Because if we going to spend about per mile on charging on charging call it 5 cents for cleaning and wear and tear call it 10 cents for insurance I think that insurance

[11:41] will be provided by Tesla. I don't think other insurance carriers will be able to underwrite it. And Tesla should honestly take the profits on it, because I think the cars will be very safe. That leaves you with a potential growth profit of 55 cents per mile. So what I wanted

[11:57] to do is figure out what would it cost on an interest-only loan to buy 10 of these. An interest-only loan on 10 of these for $35,000 a piece, which is what I think the price will end up being, I get there by adding about 35% inflation to the $25,000 car.

[12:12] So you still have the $25,000 car, but with 35% inflation after all the COVID money printing, you're at about $33,750. So call it $34,900 is what these things will sell for.

[12:24] That would be my guess, and it would honestly be reasonable to say it's an inflation-adjusted $25,000 car. But anyway, these vehicles need a cranky money. If I go finance these at 7% interest, hopefully rates come down, right?

[12:38] This is why we argue Tesla does have some interest rate-sensitive risks to it. But now, my unit economics are such that, on an annual basis, my fixed costs on that interest-only loan are going to be about $20,000.

[12:50] So that $18,300 for the loan plus FSD for each of those vehicles, which is a subscription, you know, call it $20,000 a year, so divided by 12, add that to the cost of interest.

[13:02] That's with no principal pay down, right? No principal pay down. Oh, look, pee-pee. With no principal pay down, $20,000 in cost.

[13:14] My potential gross profit after these sort of variable costs is about 55 cents, assuming a dollar per mile, right? I've got to break, now I've got to cover my fixed costs. So to cover $20,000 of fixed costs with gross profit of 55 cents,

[13:29] I need to drive 36,000 miles. That's per month. That means per car I need to drive 3,600 miles per month, which is an average of 121 miles per day,

[13:41] which sounds really reasonable if they can operate 24-7, but they can't operate 24-7 because we've got to charge and clean them. And this is where we get to an average Waymo ride per day is our only comp right now.

[13:53] Maybe Teslas will be a whole lot more efficient. But our average Waymo comp right now is that an average Waymo does about 18.4 rides per day. Now, in fairness, maybe that's because it's in a city. I don't know. But that's dangerously close to the break-even that we have here.

[14:08] 20 rides. at an average Uber distance of six miles on average. Average Uber ride is six miles on average, which means our break-even is about 20. We start making money after 20 rides per day.

[14:21] So we need to be more efficient than Waymo to start profiting. We need to beat Waymo efficiency. But even if we start beating Waymo efficiency, which I think we will do,

[14:33] if this compresses, this $1 per mile compresses, I mean, it could be more. It will be more initially. But if that compresses, that's going to kill our ability to generate volumes on private cap sales.

[14:47] So that suggests that in the long run, for volumes to really keep growing at Tesla, you need to have a business case that makes sense for a business owner.

[14:59] And the variables for a Tesla investor to look at are going to be what can be charged on a robo-taxi per mile, A dollar per mile is what you want to average out at, at bare minimum.

[15:12] The more you grow above a dollar, the better it is. The more you fall below a dollar, the more of a risk you end up with capital fail. Missing. And then when we actually look at the loan payback period, you know, this was without principal paydown,

[15:28] we need to beat the averages that robo-taxes are doing today. I think we can, but even if we double this, now we're only getting back to paying back principal. Now, in fairness, you've got a tax benefit, and that's the big element that I think adds to margins at Tesla.

[15:43] I think Tesla could potentially get back to 30% growth margins because they're able to sell this vehicle to somebody who's able to write the whole thing off as a business expense, thanks to Donald Trump's capital expenditure of being 100% taxable.

[15:56] That's where the real potential is for Tesla. But it does give us some more clarity on the risks. Risk number one, benefit per mile. Risk number two is even if we get to this unboxing strategy,

[16:08] how many of these can we actually sell? How many fleet operators can we actually sell to? Or we just open this up to consumers, right? Everybody wanting their own. That might be the better strategy. But then, can you sell it for $35,000?

[16:23] Probably not. Because it's a two-door vehicle with a two-seat setup. And that's not going to be functional for most people. So it's going to be harder to sell that to a consumer. And that's where we get to the timing issue,

[16:37] which is what was brought up right here in the interview with Art Spratt. So let's listen to this. A book that leads two people in a taxi. And you're really into cars.

[16:49] I mean, are people going to be into two-seat robo-taxis? Does it matter how many seats there are? I remember this is the idea of owning a car is what we talked about versus renting a car.

[17:02] Just as we just said, hey, are people going to want to buy a two-seat car? Probably not. Will you ride in one? Probably. But let's listen to Brett. I mean I think from the general idea and this is something that Sam Korf who our batteries analyst and I have been discussing a lot is that probably optimum strategy for Tesla for robo is a two vehicle

[17:30] It might be four seats, but it's really a two-seat vehicle. Nailed it. Because if you can deliver robo-taxi service to your vehicle, you want to minimize your manufacturing costs and maximize your unit volume,

[17:43] so you can basically maximize ride liquidity, like maximize the number of places marked where they're on service and maximize your data intake rate. And I think this is back when they were still thinking about

[17:55] Tesla owning all of the fleet. I personally think they subcontract it all out. All the low margin stuff goes to the subcontractors. Let them take the tax benefit. Tesla takes the profits up front.

[18:07] That's my thesis. It's not true that a four-door vehicle costs twice as much as a two-door vehicle, but it definitely costs more than a two-door vehicle. All right, so it seems likely to me that the platform they developed for RoboTaxi is a two-door car.

[18:23] The issue is if somebody doesn't believe that RoboTaxi is possible and they find out Tesla's next-gen $25,000 car is a two-door car, traditional analysts will conclude, well, there's no market for two-door cars.

[18:38] Right, this is an important warning because it aligns with what I just said. You need demand from fleet operators who are going to be business owners, who are going to look at the unit economics of a global cap.

[18:52] If the unit economics don't make sense, your buyer pool diminishes. And that's where then people say, okay, well, who's going to buy the cyber caps to use themselves? Well, some people will for fun or sport or whatever, and it's cool.

[19:05] But most people won't. So that's an important warning. So Bob Gutt was right about $2. door. And I think he's right about this warning as well. It has to make sense for the end buyer.

[19:18] It's not like the Honda Civic. It's two doors. That's a four-door vehicle. And they'll conclude that this will actually kill the sales demand for the vehicle if you don't believe in robo-taxis.

[19:30] Now, one, I think that every market that Tesla's entered into has been a market that didn't actually really previously exist. And then they launched the Model X even. like SUVs at that price point were not a thing, right?

[19:44] And actually they moved people off market into it. There are kind of, and expanded the category meaningfully, like there is like a neighborhood electric vehicle category that's lying nascent.

[19:56] There's like smart cars that, you know, are fine except if you're driving around with a bunch of SUVs, you might get like pancaked by somebody on the road. And I suspect that kind of like even ex-RoboTaxi Tesla

[20:09] launching a $25,000 vehicle that has two doors will actually sell quite well. But, you know, our expectation is Drogo Taxi will commercialize basically this year

[20:21] or next year. That was another warning. This was filmed at the beginning of 2024. The argument was that it would commercialize 2024, 2025. We're at the end of

[20:33] 2026, man. And we're not at commercialization yet. That's another warning. so it's not just the end buyer it's time commercialization event will you know

[20:45] transform the economics of Tesla and then it will on margin specifically completely agree that it will transform the economics I think they think because Tesla's going to keep maintaining and driving the car I think because

[20:57] they're going to sell them you know mean that that kind of mass market type vehicle is probably not even being sold into individuals in a material way it's being acquired by operators who are trying to operate up as part of the regular taxi network.

[21:11] And then, with even a two-seat vehicle, you can meet more than 90% of the demand for kind of regular taxi and ride-tail. I totally agree. And that is exactly where the EU, that economics have to make sense.

[21:27] And I think they might, hopefully, as rates come down, but you don't escape the rate sensitivity of this, because I think a lot of fleet operators are going to end up financing us. So there's a lot to unpackage here, but I think the bottom line is when we go to a forecast on Tesla,

[21:42] when we say that the bull case for this stock is $775 per share at the end of 2030, that does not include optimist revenue.

[21:54] I purposely set that to zero, and I purposely set semi-trucks to zero. I'm not saying they're not going to make money from those things. I'm just trying to be conservative. You know, if I go put a billion bucks, well, I'll put $1.5 billion back into here.

[22:09] Hold on a sec. Let me, let's just, I don't know, let's take a number over here. If I go back in, I throw $1.5 billion in here, and I throw $1.5 billion into the Tesla lot.

[22:22] Let's see if I can make it really quick. You know, now all of a sudden, we could get to $800, right? So that's not actually that big of a mover, because a 30% margin on this top line, that's the big mover.

[22:34] The cyber cap is where the dollar hollas are for the next decade. And the pros and cons are clear. The economics have to make sense. The cost per mile has to make sense. And it all starts with number one, which is edge-based mapping.

[22:49] Always start by edge. Join our advertising. These things that you told us here, I feel like nobody else knows about this. Will Smith, Tyler, advertising and Seattle Coats. Congratulations, man. You've done so much. People love you.

[23:01] People look up to you. Kevin Pasquale, finance analyst and YouTuber. Meet Kevin. Always great to get your take.