We unpack Tesla’s Q3 2019 results including their record production, record deliveries, positive free cash flow and return to GAAP profitability. We also discuss what’s going on with Model S & X and their future, the expected exponential growth for Tesla Energy, the soon to be unveiled Tesla Cybertruck, Model Y being ahead of schedule, the importance of Gigafactory 3 in China, and the latest roadmap for Full Self Driving.

If you are planning to buy a Model 3, S, or X please consider using this referral code to help support the show and get 1,000 miles of free Supercharging, http://ts.la/caleb80748

Transcript

Caleb:
Hey Mike.

Mike:
Hi Caleb. How are you doing?

Caleb:
I'm doing well. How about yourself?

Mike:
I'm doing well. We're so rusty. This is our second attempt at starting the show.

Caleb:
Nah, it's the first attempt.

Mike:
There was a lot of amateurish crosstalk, crush talk? Yeah. This one, this one's going much better.

Caleb:
Yeah, but we are excited because it's a Q3 2019 episode talking about quarterly earnings. Before we get into that, Mike, what are you drinking?

Mike:
I am being kind of lazy tonight. I'm feeling a little bit of a flu coming on and everyone knows the best thing for a flu is whiskey. I'm just having some a Yoichi. It's a Japanese single malt from I think Nikka.

Caleb:
Okay.

Mike:
It's very nice. It's a little bit PD, but not like Laphroaig level PD. Just got a little hint of peat and it's very mellow and it's soothing my throat right now. I think it's going to make me feel better.

Caleb:
On the rocks or straight up?

Mike:
Just neat in a glass. Just, yeah, I just opened up a bottle and poured it into a glass.

Caleb:
Heavy glass?

Mike:
It is a hard strong glass with a gold rim.

Caleb:
Ooh.

Mike:
It's a single rocks glass.

Caleb:
Nice.

Mike:
One of my many glassware purchases from Umami Mart. It's a bit of a problem. Anyways, how are you doing? What are you drinking tonight?

Caleb:
I'm drinking a martini. I thought we could have something a little classy for this financial episode.

Mike:
That is classy, like international man of mystery classy.

Caleb:
Yeah. It's some St. George Terroir Gin. It's quite deep and strong, junipery. Then it also has just the regular St. Martin. Oh no. Just the ... What's it called? What is a green bottle-

Mike:
The martini?

Caleb:
Vermouth.

Mike:
Yeah. Martini Rossi.

Caleb:
Just Martini Rossi. Just the white vermouth.

Mike:
Did you brine it up? You usually-

Caleb:
Oh yeah. It's dirty. Yeah.

Mike:
Oh yeah. I don't know. To each his own.

Caleb:
I just feel like that's the way to go. I like it.

Mike:
It is an accept ... I mean, it's not like you're making it with vodka or something like some sort of animal. It's an accepted thing, but yeah. To each his own. Cool. What do you garnish it with?

Caleb:
Just an olive stuff with pimento.

Mike:
Right, of course. Yeah. I don't usually-

Caleb:
No twists.

Mike:
... deal with the dirty martinis. Cool. All right, well, we are just a well-oiled machine here.

Caleb:
I'm ready to jump in.

Mike:
Yeah, let's jump in. Tesla is still a company that didn't go bankrupt in Q3?

Caleb:
No, far from it. They had a really good quarter. Coming into the quarter, the expectations were, I would say a bit muted. The last quarter wasn't very great, pretty big losses. And so, people-

Mike:
Any SEC violations or was it a quiet quarter?

Caleb:
It was a very quarter. I would say. Everyone did a very good job of not getting into any hot water on Twitter. It kind of was just a little bit slow. I think people were not expecting much, but ultimately they had some records in multiple categories. Off the top, record production of 96,000 vehicles, the most vehicles they've ever produced out of their factories. They have a record-

Mike:
Wait, sorry, factories or factory?

Caleb:
We'll get to that, but they did start doing some test production in Shanghai Gigafactory. They made some cars in China in the quarter, but they did not count towards that production because they are not sellable yet.

Mike:
Okay. So, the 96,000 came out of Fremont?

Caleb:
Yup. Then they delivered 97,000 vehicles, so pulled a few vehicles ahead from previous inventories, which is great. That's another record, record deliveries. They had 6.3 billion in revenue, but the revenue number as expected was a bit lower than previous quarter, the previous quarter, $500 million less. Off the bat what that tells you is they had record production but not record revenue. I don't think it takes a math genius to realize that means that they didn't generate as much money per vehicle as they have in some previous quarters, so we'll get to that.

Caleb:
They had a positive profit. They had a profit margin of 18.9%, which was higher than last quarter's 14.5 so more efficient. We'll talk more about that. They had $1.19 billion in gross profit compared to 921 million last quarter. That's just the revenue minus the cost of the vehicles and the production cost that go into them as well as their solar components.

Caleb:
CAPEX, $385 million on capital expenditures, so think work for the Fremont factory, for Shanghai Gigafactory, as well as superchargers. That was over $100 million more than the previous quarter, so more investment this quarter than last quarter, and 930 million in OPEX, everything it takes to sort of run the business. A lot of the staff, R&D, salespeople, the stores, service centers, things like that.

Caleb:
What's interesting is this was down almost $200 million from 1.1 billion last quarter, which is really intriguing because OPEX typically grows over time. Do you hire more people?

Mike:
Did they have layoffs?

Caleb:
This was partly the result of last quarter, they had layoffs or slightly before last quarter and last quarter, I think they had over $170 million in restructuring costs in terms of like early severance packages for folks. This quarter, the reductions actually started showing up in the OPEX line. So yeah, primarily due to the layoffs earlier and probably some reduced hiring. That ended up resulting in 371 million in operating cash flow and a total of 143 million of GAAP profit. No shenanigans profit, real generally accepted accounting principles profit of 143 million, which compares to a loss of 408 million last quarter, so over a $500 million swing to the upside and less revenue. So really, really positive. We'll talk a little bit more about how that all happened and whether or not it's sustainable.

Caleb:
Ultimately also to your earlier question, not anywhere close to bankrupt. They have 5.3 billion in cash, up $383 million quarter over quarter.

Mike:
Where does the extra cash come from if they made 143 and got profit but cash on hand went up 383?

Caleb:
Yeah. It's almost all from the $370 million in operating cash flow that came up. Basically, they generated more cash than they actually had in profit, primarily due to depreciation and amortization expense, which you have to pay for in GAAP, even though it's a non-cash expense, as well as their stock-based compensation. The depreciation, which is basically the factory, equipment decreasing in value over time because at some point, it will become inoperable and you'll need to replace it. And so, when you buy it, you depreciate it over time to try and account for that on the books and then-

Mike:
All the laptops everyone has.

Caleb:
Yup. Those also get depreciated. That's a non-cash expense. You don't actually pay for it out of your pocket every month. You already paid for it upfront but you account for it in your income statement every quarter. Then also stock-based compensation, it doesn't cost cash to give people stock options, but it does have a real economic cost to shareholders. And so that also reduced their GAAP profit. Yeah, that's how they were able to increase their cash even though they only had 143 million of GAAP profit.

Caleb:
Yes. Elon started the call saying Q3 was obviously a very strong quarter and we had record deliveries. We're able to make great strides in controlling our costs and we shifted back to GAAP profitability while also generating free cash flow. Again, this would not be possible without each employee doing their part to reduce costs. Our operating costs is now at the lowest level since Model 3 production started.

Caleb:
To me this was sort of the most surprising part. I was not expecting them to be profitable this quarter. I think one of the overarching themes for this earnings call as well as the slides that they shared with investors was around cost controls, margin expansion and sort of just general operating efficiency, which is not something we've typically associated with Tesla's model. They have really been and I think we've been quite pounding the drum that there really have been for the past year or two an investment mode. Ever since they unveiled the Model 3, they had begun putting plans in place to invest quite heavily to make the Model 3 in quite a large capacity and certainly going from 7500 to 100,000 Model S and X to what they had claimed to wanting to get the half a million Model 3s a year, we knew that was going to be a big investment.

Caleb:
And so, when they had losses and negative cash flow, we weren't necessarily very concerned, but now they're at a place where they've been relatively steady with their production of Model 3 and they've sort of teetered back and forth between profitability. Now they are back to profits on strong deliveries but not necessarily the most revenue. I think the fact that they were able to reduce their operating expenses as well as improve margin is a really strong sign that they have really been paying attention to the profits.

Mike:
well done.

Caleb:
One of the other things that's pretty interesting here too is that they expect to have positive quarterly cash flows, free cash flows going forward, as well as GAAP net income. Essentially, we heard the statement from Elon in the past that they were going to be self-funding and continuing to have free cash flow and GAAP going forward. They've now tempered it a bit more as well saying except when they're ramping and launching new products to give themselves a bit of cover for that, those incidences. But certainly, as they have more production capacity, the ramping of any one vehicle won't be as big of a hit on their overall operations as the Model 3 was. I think it's good to hear them reaffirm that they do plan to be cash flow positive and GAAP profitable in most quarters going forward. Also, they intend to be self-funding. They don't expect to need to go back to the capital markets.

Caleb:
We'll see. Q4 should be quite strong. Q4 has typically been quite strong. Q1 are typically their worst quarter, so that'll be the one to really, really pay attention to. That's two quarters away so we'll deal with that in a little bit.

Mike:
Yeah. Well, and with the lower price Model 3 too, it's probably more likely to end up underneath the Christmas tree than in Q4.

Caleb:
That's true. Q4 has been really, really good for their deliveries. I think this year, Model 3 will be quite a popular vehicle as a gift. It's funny, there are a lot of people who give cars as-

Mike:
I've never known anyone who does that. You see the commercials on TV where someone comes out and there's a big giant bow on top of a car and it's like, I don't know. I always thought it'd be a little miffed if someone bought me a car because it's something you want to pick yourself, right? You have opinions about like what model you want, what it looks like, the options.

Caleb:
I think it may be that it's just that's when you plan to do it and people go out and do their group purchase with their family or their significant other for the holidays and sort of treat it as a holiday gift. Also, because so many people's bonuses are annual, they tend to fall towards the end of the year. I think it also has to do with people being a little bit more flush at the end of the year, even in light of having higher spending for the holidays in terms of-

Mike:
Yeah, it's great. You'd think that would make Q1 a little higher though, right? Because by the time you actually get your bonus and get ready to spend it, like you're probably into January.

Caleb:
Pre spend baby. Pre spend that. Haven't you seen Christmas vacation? The pool.

Mike:
That's right.

Caleb:
That is a classic. That is not a cliché. That is a cliché for a reason and it's not a false cliché. Getting back to it. Sorry. One of the other things that Zack, the CFO mentioned on the call was how their automotive gross margin improved quarter over quarter about 2.8 percentage points. When you take out the regulatory credits, which is sort of the most apples to apples comparison since the regulatory credits that they sell to other automakers aren't always consistent and sort of distort the true profitability of the automotive segment, so 2.8 percentage points, 280 basis points is quite a lot of improvement in just three months. The way they described it was essentially they had higher production of S and X, which are higher margin vehicles, which that makes sense. They also had more production of Model 3 so that enabled better fixed cost absorption, which is a finance speak for the cost of running the factory gets split out over more units of production.

Mike:
Was the higher production volume on S and X, was that expected or is that an unexpected bump?

Caleb:
It was anticipated last quarter primarily because they made a transition to a new powertrain called Raven. And so, they had been winding down production and inventory of their previous powertrain in the last quarter. They had lower production last quarter because they wanted to drain their inventories and not build up as many ahead of this launch because they knew it was going to happen, which happened last quarter. Then also as a result, the new Raven update came out and so that increased demand for S and X. And so, this quarter they were able to fulfill a bit of that. Apparently, they even have more pending demand for this Raven update. And so, they think Q4 will be even more production for S and X than even this quarter.

Mike:
Interesting. At least in the short term, the Model 3 is not really destroying the sales of the S and X. They are actually making more than before?

Caleb:
They're making more than they have in the past couple quarters, but actually year over year, Model S and X are down, how much is it? Thirty nine percent.

Mike:
Oh wow.

Caleb:
Year over year, down 39%. It most likely to me, I would say that as a combination of cannibalization, so folks who had been buying Model S primarily who were stretching to get it, who now would just choose a Model 3. Then also, there's still this pent up feeling that Model S and X are due for more significant updates to bring them in line with the more modern interior of the Model 3.

Caleb:
And so, I think that's still an overhang on people wanting to go out and buy the Model S and X and mass because the current options for high-end electric vehicles still are not that compelling, especially if you were considering a Model S or X. I don't really think it's competition at this point. I think it's more self-inflicted. We'll talk about Model S and X a little bit more in a few moments.

Caleb:
Then they also were able to bring down the labor hours per vehicle as well as costs like warehousing, logistics, delivery and import related items, as well as reducing the material costs and primarily through like negotiating better with their suppliers for better prices.

Mike:
When you say they reduced the labor hours per vehicle, did they reduce the total hours per vehicle or so, are there fewer steps involved do you know or is it that more of it is automated?

Caleb:
That they are, yeah, reducing the total amount of time that humans are working on the vehicles and that is both through more automation of certain steps, but primarily it seems as if it's coming from the reduction and combination of parts into subassemblies or individual parts of sort of continual redesign. There are many parts that now have been combined to single parts that are easier to assemble and faster to put in. Then also as a result, like just don't take as much time to make because there's only one part instead of three or four. And so, it just takes less time to manufacture those as well.

Caleb:
The Model 3 you buy today will have different parts, fewer parts, lighter parts than the Model 3 you would have bought two years ago. That is just continually being worked on. Jerome, who's the head of automotive, was also saying how there's just like this continual drum internally where every single line item of cost, whether it's the manufacturing, the labor, warehousing, logistics, it's all being looked at and just continually refined and refined and refined.

Caleb:
I think that this is finally a place where the scale is large enough with Model 3 as well as we'll talk about with Model Y that this effort of a very small percentage point improvements actually makes a very large material impact where Model S and X, it just hasn't made that much sense. Model S and X are much more labor intensive. They don't have as much automation because they're just never going to be very high-volume vehicles where Model 3 could be over a million vehicle a year program. They still have quite a while to go until they get to that level of production.

Caleb:
Everything they can do now to keep pushing those profit margins higher helps the business. It's been good to see and I think it also indicates that as Elon has been saying that the Model 3 is in a pretty good production place. We're not talking about, oh my god, they're not able to produce enough Model 3s or we're not like counting Model 3s in the parking lot anymore to see if their production is working and we're not reading memos from Elon saying, "We really have to rally to get a couple more thousand vehicles per week." Or certain parts of the line are at peak capacity. Like we're not playing those games anymore. It's been another quarter of solid production.

Caleb:
If we look back a year ago, they were producing half as many vehicles per week as they are now in the same factory. They have really sort of stabilized Model 3 production. I think that's a really good thing and also kind of easy to forget how concerned everyone was about production of Model 3 even just a year ago.

Mike:
Yeah.

Caleb:
Yeah.

Mike:
Actually, I saw one. I haven't really been looking for quite a while, but I happened to walk by one that had new temporary plates on it. I just looked at the VIN and it was 500 and I don't know, and then I'm thinking, I'm not sure if it's 527 that I saw or 572 but anyways, up over half a million.

Caleb:
Yeah.

Mike:
Which is pretty cool.

Caleb:
One of the other crazy things is that the demand is still strong. Like I think it's crazy ... Someone in my apartment complex just got a Model 3 and there's only maybe 15 units here and I was the only Model 3 person in the parking lot and now there's a new one and it's brand new. It's a dual motor performance and it showed up a couple of weeks ago. I was like ... Like it's just so interesting to remember that people are still just now getting them and the demand like is still growing.

Caleb:
Elon said on the call that our global order rate remains strong and continues to increase. Despite increases, the production levels, our order backlog has been growing and the quarter to date orders referencing Q4 are significantly higher than at this point in the last quarter. And so, even though we've made, they've made 3, 400,000 Model 3s already, it continues to grow. Demand continues to not only be able to sustain this level, it's actually the backlog is increasing. And so, they actually need more capacity.

Mike:
Yeah. They've probably gotten out of like the Tesla fans who are looking to get Teslas and maybe they're just now into like people who are just on their upgrade cycle for their car and like their lease is up or they want a new car and just looking at the lay of the land, they decided on a Tesla.

Caleb:
Yeah. Elon even said on the call that someone had asked like, is word of mouth strong enough to continue demand for the capacity that they have? Elon said, "Yeah, that what we're seeing is that word of mouth is more than enough to drive our demand in excess of production. We have no plans to advertise at this time. At some point in the future we may do advertising, but not in the traditional sense."

Caleb:
Just again, on the earnings call in front of investors saying that demand is in excess of production. I don't think we're going to see big pile ups of Model 3s and the big concern that, oh, people see Model 3s in parking lots every once in a while, that seems to be more of a logistics challenge than anything else.

Mike:
It sounds like we're also not going to look forward to a Tesla Superbowl ad this year.

Caleb:
No. Very, very unlikely. Yeah. Again, just sort of to recap the financials. Revenues were down 1% quarter on quarter, but GAAP income was up 136%. The only way you can do that is through efficiencies. And so, automotive gross margin was up to 22.8% from around 19%. Gross profit was up 20% quarter on quarter and operating expenses were down 15%. More gross profit and less operating expenses leads to that 136% increase in GAAP income.

Mike:
It's just that simple.

Caleb:
It's just that simple.

Mike:
It was our MBA.

Caleb:
Also, no big changes to debt. The only reason debt kind of increased was that this year compared to last year this time, they have over three times as many vehicles that are being leased. And so, the lease, the way that leasing works is that Tesla has to keep the debt for the vehicles on their books until they get paid back by the leasing companies via the customer. Yeah, no real concerns about increasing debt or anything as a result of hitting these results because debt doesn't necessarily hit the income statement except for interest. Because interest rates are so low, it doesn't really have a big impact. A lot of companies are taking on a lot of debt. Tesla has not added to their debt in any meaningful way this quarter, just as a point of reference.

Mike:
Cool.

Caleb:
Yeah. On the operation side of things, that production rate from Model 3 of around 80,000 vehicles works out to 6000 Model 3s per week, assuming a 12-week quarter. Remember like a year or two ago, Elon was like, "We're going to get the 5000 Model 3s a week and then like in a few months, we'll get the 10,000." They're still more close to 5000 per week than they are to 10,000.

Mike:
Are they still using the big tent in the parking lot?

Caleb:
Yes. Yes. They are still using that as part of the production process. And so, this is a little bit ... I'm not sure. I'm not sure if it's concerning or if it's a demand issue. We were just talking about demand, but essentially, it's not clear to me if they underestimated what Fremont's capacity actually ... Sorry. Overestimated what Fremont's capacity actually was or in managing their profit margin and the price point are not yet at a place where they can lower the price enough that would push demand high enough to support closer to 8 or 9000 Model 3s.

Caleb:
Or the alternative theory is that they expected that they would be ramped faster and that they would have gone to 10,000 Model 3s per week in the US anywhere, way before Shanghai was ready. Now that the Gigafactory Shanghai is so far ahead of schedule that they've decided to not push very hard in Fremont on production but just focus on margin improvement and wait for the Shanghai factory to come online because they know they have more capacity and they need to make cars there.

Mike:
My money would be on the overestimating.

Caleb:
Yeah. I think they definitely overestimated how much that production line could do because they were really talking a lot about how the new Shanghai production line is like a next generation version. I thought the first.

Mike:
They're overestimating that one now.

Caleb:
Yeah, probably.

Mike:
I mean, we never got the coast to coast drive that we were supposed to get to, right? I feel like-

Caleb:
I've got updates on that.

Mike:
Okay. It's just ... Yeah, they ... Yeah, a quiet quarter is nice because in general that means there was less overpromising I guess.

Caleb:
Yes.

Mike:
That's just a chronic problem.

Caleb:
It was definitely ... There weren't any major milestones targeted for this quarter except for V10, which they did do about a month or two late, but it did come out.

Caleb:
If you look at the trailing 12 months, which is a nice way to sort of smooth out the results, because you can sort of smooth any of the perturbations for any particular month or any particular quarter.

Mike:
Yeah.

Caleb:
If you look at vehicle deliveries over time, sort of 3Q 2018, the previous 12 months, they'd done around 190,000 vehicles and this year, they're at around 350,000-ish. Year on year, quite a lot more vehicles for the trailing 12 months. Then even if you go back another year, they were just under 100,000. Last year was really the big step change in Model 3 production happening and this year, only around a 50% increase in production, still a lot in terms of the absolute number, but nowhere near the levels of change. And so, I think that's also partly why it's not as crazy as the absolute percentage changes or just not as big. As a result, we don't have as much change in any month or any day. It leads to more stability in the company.

Mike:
That sounds like a good thing.

Caleb:
I think so, especially for Tesla. If you look at the net income trailing 12 months, it's this interesting upside down, I don't know, wave where in 2016, their trailing 12 months profit was around negative 1 billion. Right now, it's about negative 1 billion for the past 12 months. If you go back to like midyear 2018, they were down at like two and a half billion losses. They're kind of getting out of the hole in terms of losses. That really was the peak of Model 3 ramping.

Mike:
Yeah.

Caleb:
That's what they kind of talk ... That's what they mean by when they say when we ramp new vehicles, we're going to have some potentially negative quarters and you can see that in their results here. My point is just that it won't be as pronounced because Model 3 will be able to absorb a lot of that. Also model wise, it's not going to have as much trouble. Fingers crossed.

Mike:
Well and just from listening to you too, from a marketing perspective, I think saying you have a profit of negative 1 billion sounds better than a loss of 1 billion. You think?

Caleb:
I don't know. That's interesting. I wasn't trying to spin.

Mike:
No, no. I think you said it like two different ways, which are the same thing. Yeah. I guess it's-

Caleb:
A profit of negative 1 billion sounded better than a loss of $1 billion.

Mike:
It's a little Orwellian, but it sounds ... Yeah. It has a more positive connotation to it I guess.

Caleb:
I should really push that for some of the technology companies that are out there right now.

Mike:
It's more of a good news, bad news kind of thing. We have profits of negative $1 billion.

Caleb:
Yeah. Just that negative sign. It's really small. Like you could just make it a little tiny little minus sign.

Mike:
Or put it in parentheses like in the spreadsheet.

Caleb:
Yes.

Mike:
It just sounds like you're whispering your profit.

Caleb:
Yeah. Just make it red. Yeah, easy. Profits of 1 billion red.

Mike:
This is probably why we're not accountants.

Caleb:
No, no. I think we would get ... I don't think they really like marketing people becoming accountants, kind of not the career trajectory you want to see.

Mike:
No. We've had a little too much problem with creative accounting really.

Caleb:
Exactly.

Mike:
In our economy as of late.

Caleb:
Yeah. It's not really a left brain kind of situation. On solar and storage, an area that still isn't that critical to the company from the revenue point of view and contribution margin point of view. We don't really talk about it too much on these calls and recaps for that reason. However, I wanted to spend a couple minutes on it here just because of a few important points.

Caleb:
One, they had all-time records storage deployments. This is battery packs, Powerwalls, Powerpacks, and soon Megapacks. The three megawatt hour battery packs are going to start in Q4. That's going to drive megawatt hours up a lot. They did 477 megawatt hours of deployment, which was a record. Then also, the solar business grew 50% quarter on quarter. Those are the just solar panels that they do for residential and for commercial customers, still way down from when they purchased Solar City, but the model is still primarily around purchase versus leasing.

Caleb:
The other really big thing that occurred on the call and then in the subsequent days, which now because we get to record a little after we can speak about is they unveiled the new solar roof. This is V3 of the solar roof. I mean, the original solar roof was unveiled over two years ago, Mike. Do you remember when they-

Mike:
I feel like that's been a while. Yeah.

Caleb:
Remember we did that call. We did that recording and they had gone to Universal Studios.

Mike:
Yeah. There was just like a house there and it-

Caleb:
Yes.

Mike:
Yeah.

Caleb:
They did it in the back lot where they filmed Desperate Housewives and it was like, oh boy, this new thing is coming from Tesla and the Model 3 wasn't really even in production yet and all this stuff. It was super exciting. Here we are two plus years later and it still is like on a handful of roofs, max and I think they really underestimated how hard it was going to be to do both the technology and what it sounds like actually now has been the actual installation process and sort of how do you make a custom roof a product that can be high margin versus like a shingle roof, asphalt shingle. They now seem to have gotten the price down 30, 40% over the V2.

Mike:
It is them getting the price down or just the kind of constant falling of the solar cell prices? Are they just riding that wave down or are they doing anything?

Caleb:
No. They did a lot to reduce the number of pieces that go in. They made the panels like each individual tile, like almost double the dimensional space of the old ones, reducing the amount of site work, drastically reducing amount of time it takes to install. There's a lot less labor hours involved in installing it. We might do a deep dive on the call that they did about this, but essentially, solar roof-

Mike:
Have they shown photos of them?

Caleb:
Yes. They have them on their homepage now. Yeah. The ones they're doing initially are the black glass ones, so they look more like regular.

Mike:
Yeah. Well, I'm wondering because the things that you're saying, like the whole selling point of these solar roofs is that they look just like regular roofs, but then as you start to do things like double the size of the shingles and change the materials, you could easily kind of iterate your way away from looking like a normal roof and starting to look like a solar roof.

Caleb:
Yeah. They stayed ... I don't think there are many roofs that look like this currently in the current design. They still claim that they're going to do like a more slate type one, another one that looks more like a terracotta.

Mike:
Because terracotta was the one that on Wisteria Lane, right?

Caleb:
Yeah.

Mike:
The one that they demoed.

Caleb:
Yeah. I think they do still plan to do that, but they are going to keep ramping, ramping them up. The other kind of important thing is that they plan to partner with roofing contractors in six months or so to actually certify them to install the solar roof. They hope to be scaling up to like a thousand roofs per week worth of capacity by the end of the year.

Mike:
Oh. So, they've learned their lesson about estimates then in there. They've-

Caleb:
It's not 10,000 at least, yeah, but a thousand roofs. Each roof is about the price point of a car.

Mike:
Yeah.

Caleb:
That's actually going to be quite significant. This is what Elon was saying on the call that I wanted to reiterate was that someone asked, "Of everything that's happening in the Tesla business, what do you think is the thing that's least appreciated by investors?" Elon said, "I think there's generally a lack of understanding or appreciation for the growth of Tesla Energy, as Kunal was talking about who leads the Tesla energy. In the long term, I expect Tesla Energy to be of the same or roughly the same size as Tesla's automotive sector of business. This is the most underappreciated group. I think it could be bigger, but it's certainly of a similar magnitude to Tesla Automotive. Meaning if you take Tesla' solar plus battery stuff, Tesla Energy, I think that's the least appreciated element.

Caleb:
Again, like we keep hearing, we keep having Lucy move the football, that Tesla energy is going to be this massive equal part of the business and it hasn't materialized for over three years.

Mike:
California is providing a pretty good opportunity here with all of our power problems. We should ... Tesla should be working to micro grid this and create more sustainable small scale energy.

Caleb:
Apparently, he said ... Someone did ask about that, have the blackouts in California cause an increase in Powerwall and solar. He said, there has been a slight blip, but he said just in general, the overall demand has really been hinging on people wanting the solar roof. And so, they feel like they've just had a huge backlog of that. One of the reasons he claims that things haven't progressed as quickly, and I think this does make sense, was that he was saying, and I quote, "As part of it, like for about 18 months, almost two years, we had to divert a tremendous amount of resources. We had to basically take resources from everywhere else in the company and apply them to Model 3 production, fixing the Model 3 production ramp and simplifying the design of the Model 3. So for about a year and a half, we unfortunately strip Tesla Energy of engineering and other resources and even took the cell production lines that were meant for Powerwall and Powerpack and redirected them to the car because we didn't have enough cells."

Caleb:
He's basically now post Hawk admitting that part of the reason the Tesla solar business hasn't advanced and they haven't been able to make as much progress on solar roof is because they move almost all the talent to focus on solving Model 3 because that was an existential risk. If they didn't make Model 3 successful, the whole company would have gone bankrupt. We know that they were quite close to that multiple times with Model 3, but they've finally been able to get that to a place where they feel comfortable and therefore, they've been able to reallocate those folks back to Tesla solar.

Mike:
We'll have to put a pin in this and when the Q4 numbers come out, come back and check to see if they are getting towards that 1000 roof mark.

Caleb:
Yes. I am very curious if solar roof does move at the ramp pace they said because we should do a deeper dive before that happens so we can go a little more, because they did an entire like 45 minute, an hour call about solar roof. It could be useful to dig into that.

Caleb:
Moving on to S and X briefly, we've already spoken to that a bit. A couple more things I found kind of intriguing was that they are working to increase production of Model 3 and Model S and Model X. They say the bulk of that work involves continued optimization of the existing equipment. Then they are increasing production for Model S and X lines for this quarter in response to increasing demand. They anticipate that demand will be reflected in the S and X deliveries in Q4, which I think you have to say would be higher than Q3 sequentially.

Caleb:
Here's something that really scared a lot of people, and this was Elon. "The Model S and X are really niche. They're just really niche products. I mean, they're very expensive, made in low volume. To be totally Frank, we're continuing to make them more for sentimental reasons than anything else. They're really of minor importance to the future.

Caleb:
I think we've kind of known this for a little while that-

Mike:
Like I said, that doesn't sound terribly shocking.

Caleb:
Yeah. They've been talking about this for a while that like why aren't you pushing Model S and X? Why isn't that growing? Elon has been pointing to, well, Model 3 is our focus. Model 3 is our future. I think people expected that Model S and X would bounce back to their previous Model 3 levels once Model 3 had gotten to a more stable level like it has now, but they haven't.

Caleb:
I think what scares people who are customers of Model S and X and one of our friends has a Model X and he was concerned hearing this because he loves his Model X and he wants to get another one at some point when he's done with this one. I don't think this means they're going to get rid of the Model S and X. I think they are still going to keep making improvements. When you think about Shanghai Gigafactory, there is no plans to make S and X there. When you think about the European Gigafactory, there are no plans to make S and X there. Really the future, and Elon said this on the call, the future is Model 3 and Model Y. The pickup truck, the Tesla Semi, the Roadster are all going to be much lower volume. The pickup truck, who knows, maybe that will really be big, but pickup trucks aren't nearly as popular globally as they are in the US, so you might be able to get by with just US production for the pickup truck, especially because the price points are quite high.

Caleb:
I think it just sort of points out that the S and X were quite useful for launching the company and are nice halo vehicles. They are not going to be investing in them nearly as much as they invest in the improvements for Model 3 and Model Y. I just wouldn't expect a lot of work on S and X unless there is some new technology that they want to test out there that's easy but not lots of big massive redesigns.

Caleb:
Now, do I think they're going to roll out the screens in a new format in Model S and X at some point? Yes. It's not priority. I think not being priority is a very dangerous thing at Tesla because they usually have so many things pushing for priority and challenges happening that being four or five on the priority list might mean you really get pushed out really, really long timeframes and things that you might've thought were going to come this year come like a year later or two years later.

Mike:
Yeah. I'm sure there'll be making motorcycles and drones before they get to updating the S and X.

Caleb:
Yeah. It's a bit frustrating for folks who like those vehicles. I understand. Those are also the categories that are going to have the most direct competition from other companies right away. it's also like not the area that is that large of a market and also, they already are besting the competitors from an efficiency point of view. They don't have to work that hard to continue to sell those vehicles.

Caleb:
Moving on to Model 3, not too much more than what we've already discussed. Just that despite the decreasing average selling price, the margins are improving. This is like a really fundamental thing because typically when you sell a lower price version of your product, you're not going to have as much margin because you typically you can charge more for fancy bells and whistles that are really high margin. And so, the cheaper something is, the harder it is to make any margin on it because there's usually a lot more demand for cheaper products.

Caleb:
And so, despite having the mix go down because of the standard range and standard range plus having less initial performance models going out to international markets, the margins are improving, which means the fundamental costs and production capabilities of the vehicle are getting better.

Caleb:
The other really intriguing thing to think about to your earlier point about still seeing brand new Model 3s on the road is that, and this is kind of the shocking actually, is that of all the Model 3s ever made, 22% of them were made in this past quarter. It's been over two years of production and nearly a quarter of all of them that have ever been made were just made in the past three months. They were all purchased. They were all sold.

Mike:
That's pretty wild.

Caleb:
It's still pretty impressive to think about how many of these things are arriving and people are having their very first experience with the Tesla and exposing it to their friends and exposing it to their family members and the retention rate of Tesla customers is so high that in three or four or five years when they're ready for a new vehicle, it's going to be such a massive upgrade cycle.

Caleb:
I feel like this is really that cycle where it's like we're kind of in the, I don't know, maybe the iPhone 4 similar timeframe when it really started to take off. New people ... The thing that was so special about smartphones was that you'd buy a new one almost every year because the capacities are increasing so much in terms of the capabilities, that you got this really high hockey stick exponential. You won't see the same level for Model 3. I think for Tesla, like you are going to see an extremely high loyalty rate based on what we've seen with the S and X customers and even Model 3 customers trading up when they need a new vehicle like we've seen with some of our friends.

Mike:
Yeah. It'll be interesting to see if the sort of stagnation of the S and X deters people from upgrading their 3s. If you want a bigger vehicle or if you're in that sort of cycle of you've had a Model 3 or two Model 3s up in the future, and you want a bigger vehicle but like, will the dated sort of interior and functionality of the S and X put you off and will it feel like a downgrade? I don't know. It's an interesting question to think of people kind of working their way up the product line over the course of years at Tesla.

Caleb:
Yeah. Does it mean that with software updates, it may actually extend the useful life of your vehicle?

Mike:
Right.

Caleb:
So that what people may have been on a three year upgrade cycle if you're kind of aggressive, maybe a five year if you're like more standard, if that adds one or two years, you're actually kind of ... Tesla is hurting themselves in a way where the challenge for smartphone makers right now is that the replacement cycle has really extended because the new smartphones aren't as big of a leap as they used to be. Certainly, buying a Model 3 now from one, two years ago is better, but is it 30 or $40,000 better? It's probably not.

Caleb:
If your life changes and you need more capacity, you're going to have a Model Y to go up to, but yeah, to your point, I'm not sure I would want to buy a brand new Model S or X right now given it doesn't really feel like there's that much. I don't know. It's tough. They are better, but they're not ... To me, I would feel like there's some big downgrades from coming from Model 3. I think that's potentially something they're not fully appreciating that there might be more people who would like to step up to a bigger vehicle, slightly more lux vehicle now that they've had a taste with the Model 3 and they look at it and they're like, "I kind of like the way the Model 3 screen is like," or "I would like to have the vents like the Model 3," or "I'd like to have handles for getting in and out of the vehicle," and very simple things like that.

Mike:
I wonder if as you're saying this and Tesla updating the cars as they go, I wonder if they do any sort of like performance conditioning of the battery over time. because batteries naturally degrade and it would be reasonable for them to somewhat adjust the characteristics of the drive system to adapt to the sort of life cycle of the battery. That's something that Apple has gotten in trouble for because people say they're ... This whole like you're throttling my phone kind of thing, when from Apple's point of view, they're actually trying to make the phone perform well given the degraded battery. It's this weird thing where people are misunderstanding what's happening. I wonder if Tesla is like watching what's happening there and wondering if that's something that ... Is there going to be some sort of Model 3 class action lawsuit in 2025 or something from angry people who don't understand like battery technology.

Caleb:
I think there'll definitely be people who are upset when they realize like when they're in 500,000, 600,000 miles range that the battery isn't nearly as good as it used to be. I think that for the cars, because the useful life of the battery is so much longer than the sort of useful life of the screen and the components in an iPhone that it won't be as mismatched where the extreme is like the AirPods, where it's like after a year and a half, two years like they're kind of shot even though the speakers are fine, the plastic is fine, all these things. The bigger the battery pack for lithium ion obviously, the less kind of harsh the conditions.

Caleb:
And so, as Tesla just has it larger and larger battery packs in general in vehicles going forward, that becomes a lesser and lesser challenge because they don't have to stress them as much. Certainly, the people who've done really intense supercharging very frequently, they'll stop you from doing that and your peak 100% charge range does go down over time and so people do get pissed off about that.

Caleb:
It is going to be ... That's probably just like the biggest not maintenance, but like corollary to having stuff go wrong with your internal combustion engine car is like, things won't necessarily go wrong the same way in an electric car, but this is a whole new concept that your car won't go as far as it used to when you bought it fresh and like you have to get used to that.

Caleb:
Upgrading your battery may or may not make sense given the interior is getting a little beaten up and it'll be interesting to me to see how long Model 3s last on the use market. Will we see many million mile Model 3 is being sold at some point or will they just be junked because it's people ... Like the interior just doesn't ... It just gets worn out when you have that many people in your vehicle for that long.

Mike:
Yeah. Well, because it's not also built to like ... It's not built like an airplane seat or something. It's more of a personal vehicle, which gets into the whole problem with the Tesla network. But we won't dive into that right now.

Caleb:
We'll get into that at another time. Moving on from Model 3, Shanghai Gigafactory. I think this was definitely something that we've been tracking, but it's really great to see like more official commentary from Tesla on it because we've only gotten a few updates on this. It's really been just approximately three earnings calls where they've even been talking about the Shanghai Gigafactory because it's only been in production ... It's only been in construction for 10 months.

Caleb:
What's really exciting is that it's ahead of schedule and that trial production of vehicles has already begun. The body, the paint, general assembly, all of that is working and vehicles that are going through that entire process. They're currently working with the Chinese government to finalize a production license so they can get sign off to have these vehicles for sale. But 10 months-

Mike:
Is Tessa working with ... I'm sorry. Is Tesla working with any sort of ... I remember one of the things with Apple was the issue is about the conditions in the factory and they were ... I think they work with some sort of outside auditors to make sure that Foxconn or any of their contractors are meeting certain standards. When I hear things about like crazy timeframes as to how fast things are being built, I'm wondering is there any costs that are being hidden here or like are 10 people a week dying while making this factory? Is there any way of knowing any of this?

Caleb:
I'm not aware of them working with any labor auditors or safety auditors. That being said, due to it being Tesla and due to it being one of the most shorted stocks, that does have a bit of an antiseptic effect where if there is news that is negative for Tesla, it gets amplified. And so, my sense is, and this is just my sense, is that if there were violations of Chinese law or just general challenges of, god forbid anyone being injured or worse, that we would have heard about it. Certainly, we heard about it with Foxconn.

Caleb:
I think the scale is certainly a lot smaller. There's only a couple thousand people working at this factory relative to Foxconn, which is nearly 100,000 people I believe, a lot, lot smaller number of people. I think I would say also that China is quite good at producing and building even skyscrapers in incredibly fast times. I think probably the biggest thing is like multiple crews around the clock. Like in the US, it's even tough to get a 12-hour shift. To get a 24-hour shift with people who don't have to wait for permits and all these things, like that's doubled or tripled the pace. Something that takes 10 months in China could easily take two years to three years somewhere else. Even just by having more people working on it around the clock, it's a huge, huge increase in production time. I don't think there's anything nefarious going on there. I think if there is, we will definitely find out about it soon.

Caleb:
The other thing that's really impressive here, and I think this goes to Tesla, sort of continual improvement and sort of designing production systems, is because this is like their next new factory and they get to start from scratch instead of using the legacy systems of the production system they had at Fremont, they got to design it from scratch. As a result, the capital efficiency of the new production lines at Shanghai Gigafactory are 65% less expensive than the Model 3 production system in Fremont.

Caleb:
What does that 65% less expensive mean? They've spent less, 65% less CAPEX per unit of capacity. For every car that they can produce there that the line is designed for, they've spent 65% less on the equipment to produce that one vehicle of capacity. Does that make sense?

Mike:
Yeah. They're getting better at it.

Caleb:
Yeah, but 65% is a lot.

Mike:
A lot better, yeah.

Caleb:
For one generation, air quotes, of like production system, that's super impressive.

Mike:
I wonder how much of it is that their initial one was just very, very inefficient and there's a lot of sort of low hanging fruit to optimize out.

Caleb:
I think part of it is like, yeah. They needed a lot fewer robots. They figured out what steps are actually required. Now when they go out and buy everything needed to replicate what they're doing in Fremont now and doing it in an even more efficient way from a blank sheet, you just have to spend less money to do it. And that's super, super important because that means the margins will be better just out of the gate because you have less fixed costs to allocate to every unit of production.

Caleb:
This is a really, really big deal. Because oftentimes in places where labor costs are lower, you're not as efficient with CAPEX because you just rely on people. They don't seem to be doing that. They still actually seem to have more automation, which is kind of impressive. They're targeting 3000 units per week as a production capacity. I believe it's 150,000 vehicles per year. They didn't install like the same level of production capacity of like half a million, 700,000 like they did at Fremont. Is this still like relatively small? Also, they expect China to be the biggest market, single market for Model 3 so that's why it's here and why they've built this factory next. Even though China only represents 11% of total sales, it is going to ... It is the largest market for sedans and it already is the largest market for EVs. Currently, the Model 3s being imported have a 15% tariff. They cost more to Chinese buyers, plus there's even more incentives if you buy a locally-produced vehicle in China.

Caleb:
Shanghai is just like really, really impressive, how quickly they've put it together. It's probably and Elon was saying that he believes it's probably the most sophisticated factory that's ever been built this quickly. It's probably true.

Mike:
It's not provable.

Caleb:
Not provable but probably true. And that, yeah, they're in this trial production. Since the earnings call, we've seen even more Chinese media go to the factory and even I believe Tencent's automotive journalists has done a review of one of the Model 3s built in China and he actually has a Model 3 that was imported. He was saying that the build quality is better. The Chinese ones build quality is better. It's quieter, so it's got better seals and slightly better production improvements on the Model 3 itself. Everything else is basically the same.

Caleb:
These vehicles will only be in China, but I think it's really important that the caliber of the vehicles is as good or better than the ones made here in the US. It's funny to think that at one point in time, made in China was kind of a negative, especially in the automotive world, it was like, oh, Chinese cars are not great. But now, China is incredibly great at producing complex objects and now certainly with Tesla having so much automation, it's even more possible that these vehicles may be to better spec then the ones even made in Fremont.

Mike:
Yeah. Well, that was the way with the Japanese cars and Korean cars, right? Like they were econo cheapies when they first started and then they eventually got really, really good at it.

Caleb:
Yeah, for sure.

Mike:
The model [San 00: 59: 13] will be the same.

Caleb:
Yeah. I think it's just difficult as an American fully appreciate how big the automotive market is in China. It's really the only market that's growing and how much China is pushing for EVs and how as a dictatorial government they can force EVs to be adopted, both through carrots and sticks. They attached a bunch of photos from the factory and I'm just looking at some of these in our notes, like this is the exact same kind of stamping press that they have at Fremont, but it's like the brand new one. They have a whole bunch of the really old ones too. Seeing everything coming off, the stamping press looks a lot like Fremont. The density of robots looks just as intense as it does at Fremont or even more.

Caleb:
Just thinking like all of this equipment has been installed in less than 10 months. Just seeing these Model 3s like just rolling through and these people working on like, it's just so impressive. This whole building is brand new and you look at the one shot from the drone up above and the test cars off to the left and then like all the semi-trucks. Can you imagine if the Fremont factory had that great of access for the semitrucks to come in?

Mike:
I'm actually amazed they get the cars out.

Caleb:
I can't believe they do it at all. Yeah. Fremont is crazy. Then you zoom down a little bit and you look at the ... In January, it was a watermelon patch and in October, you've got thousands of people inside around the first vehicle they've produced. It's like this huge factory. Look at how high the ceilings are. It's like double the height of Fremont. It's super, super impressive. This is just the first thing.

Caleb:
They're already at work on a brand new building at the end of it and that new building is going to be for building the batteries. Right now, they're shipping in the packs, but they're going to be building them there. Then they're going to expand even further when they start producing model Y there.

Mike:
Do you think they're going to be hosting tours for podcast hosts?

Caleb:
I don't know. I would love to go to China to tour this factory. That would be extremely fun to do. Tesla PR, if you're listening, we would love to come toward the new factory.

Mike:
We'd love to see the factory. I love Shanghainese food. Let's make this happen.

Caleb:
Well, the other thing, it's like this is going to be the true gigafactory, right?

Mike:
Yeah.

Caleb:
Where they make cells. They make packs and then they make the cars.

Mike:
It's the dreadnaught, right?

Caleb:
This is the one that will be most dreadnought-esque for sure.

Caleb:
Moving on, we got a couple more things. Model Y, this one's really cool too. Model Y is ahead of schedule. They are going to launch it this summer instead of in the fall, three months early. Here's the thing, Y ... Okay. I do not believe Tesla would be so reckless as to set themselves up for another late vehicle on purpose when they didn't need to. Remember when the Model Y came out, we were talking about and I was like, huh, it just doesn't ... It feels like it's pretty far away. We were thinking maybe they were going to be more aggressive than getting it out in like six months, six or nine months. It was like over a year plus away and when you're like-

Mike:
If you're 2020, I just ... Because of my age I think, like 2020 is distant future for me.

Caleb:
Yeah. It still feels like the future for me.

Mike:
That's like Star Trek time periods.

Caleb:
Yes. I still feel like 2020 is going to be the most fun future year. Like moving out of the teens is going to be pretty cool. Anyways, they set that expectation that it was going to be fall of 2020 at the unveil and okay, like that maybe was like a little bit of a gimme, but it could be good for their PR to have actually hit a deadline for once. Now, they come out and say, we're going to be three months early.

Mike:
We don't think we've sufficiently underestimated that. We're going to revise.

Caleb:
Then Elon also says there may be some room for improvement there, but we're confident about summer 2020. He can't even help himself.

Mike:
It's like watching a drug addict. Yeah. It's like watching a drug addict. You're just like, can't help but go back.

Caleb:
He can't help himself from even trying to set the deadline even sooner. Also, this isn't the start of production, mind you. This is what volume production is of 1000 cars a week. They plan to be at 1000 cars a week in summer 2020 so let's give them charitably like August, which means they'd at least need a couple months ahead of time. That's like April or May of next year starting production.

Caleb:
That's plausible given that we now are seeing Model Ys driving around Palo Alto because it seems as if they have the release candidate because Elon was mentioning the release candidate and that it's great and he's really happy about it and it's better than even what they showed off. We've had many more people citing all sorts of different Model Ys. I think it's very plausible. They're already doing construction work at the factory to tool everything up for it.

Mike:
Indoors or out?

Caleb:
Indoors. The equipment is underway. It's in installation. They say they are moving faster using the learnings and efficiency gained from Gigafactory Shanghai factory design, and also that the CAPEX per unit of Model Y capacity is forecasted to be 50% lower than the current Model 3 production system in the US. They plan to utilize a lot of the improvements from what they've done at Shanghai in Model Y in the US. Also someone asked on the call if they planned, like if Model Y was going to disrupt model 3. They said, "We're not expecting any disruption or factory shutdowns as a result. We do not expect it to interfere. The body line is separate, the painting line is. Basically, we do not expect it to interfere with Model 3. We don't expect any downtime." Yeah. They said 1000 models per week, 1000 of them per week no later than the middle of 2020.

Mike:
Okay.

Caleb:
I'm excited about model Y.

Mike:
Full speed ahead.

Caleb:
Yeah.

Mike:
Have we seen what the trunk situation is on the Model Y? I feel-

Caleb:
It's a hatch.

Mike:
Okay.

Caleb:
It goes all the way up. More like Model X and Model S.

Mike:
Okay. Well maybe, maybe that will be the first one that we get.

Caleb:
I mean, look at these photos. I mean, it definitely is SUV-ey, but it's very ... It's just like a shrunken Model X with a lifted Model 3 combo. I think it's going to be very, very popular.

Mike:
Cool. Americans do love their SUVs.

Caleb:
Yeah. They don't expect any cannibalization. They pointed out that when Model X came out, Model S sales increased. I think they're wrong there. I think there are people buying Model 3 just because it's a great price point who would prefer to have this vehicle. Again, they're only going to be producing a thousand a week to begin with and ramp it, so they'll be able to see what the split is.

Mike:
It seems like the Y is coming in, too. Like the only real like competition space too, right? Like isn't like Audi and Jaguar and I think ... I'm pretty sure I just saw an electric Volvo SUV.

Caleb:
The XC40, yeah.

Mike:
Yeah.

Caleb:
This segment ... Yeah. I'm not sure if it's in the price point.

Mike:
It seems to be where other people are starting to dip their toes in.

Caleb:
Yeah. They've all focused on the SUV market. They're slightly bigger, not like full size SUVs, but the price points are like $20,000 more. It's the entry point for Model Y will be closer to 40,000. You could go and order one right now. They're using the same ordering infrastructure and-

Mike:
You can order one right now?

Caleb:
Yeah, yeah.

Mike:
Wow.

Caleb:
You can configure one with all the options the same way you can on all three. It just says, it'll be shipped to some point in 2020. I mean, I think this is a really good sign. Like they are still using over 70% of the parts are coming from Model 3. Obviously, they're very far along on the integration of it because they're running release candidates. There was an interview with Franz, the chief designer on the Ride the Lightning podcast with Ryan McCaffrey recently. In some ways, it sounded like Franz was not very excited about the car because it was sort of just like the brief was make an SUV version of Model 3 and keep it low cost.

Mike:
It's not a design challenge?

Caleb:
Right. It didn't sound like he was like enthused about it. As a spectator of the company, I think that's probably the right choice to actually have not gone nuts with the Model Y. I think it's certainly going to be very popular. I think if I get another Tesla, this is a ... I think I would get a Model Y. I would definitely want to try it. I've never had an SUV before, but the few times I've rented them, I don't know, there's something nice about a better road position.

Caleb:
Ultimately, I'm not driving super performance anyways. Like I've used my Model 3 in chill mode. You and I went on a relatively good hour drive a couple of weeks ago. We're just not doing launches all the time. It will make me nauseated. I'm not that worried about slightly worse performance driving dynamics in a slightly higher car that has more like spaciousness for the occupants. I totally get it. I understand. I know there's a lot of purists of sedans out there, but I think the functionality of this shape is better. I think people who've had hatchbacks understand that as you do and have.

Mike:
Yeah. Well, if I can't get you enormous bags of hockey equipment in and out of the back, we can't have it.

Caleb:
I mean this actually could be ... This could be your next vehicle. You still haven't purchased a SUV?

Mike:
No. Our 2007 Rabbit is still going. If she can hold on for another apparently only nine months, right?

Caleb:
Yes.

Mike:
Then we're good.

Caleb:
You really could.

Mike:
Yeah.

Caleb:
Be kind of interesting.

Mike:
It might happen.

Caleb:
You should. That would make sense.

Mike:
I'm not going to preorder, but if they are available, it'll probably happen. Because my wife likes the Model 3 and she wants an electric vehicle. She doesn't like-

Caleb:
Oh cool.

Mike:
She doesn't like destroying the environment just to get from point A to point B.

Caleb:
That's good.

Mike:
Yeah. It's definitely on my radar. If fingers crossed, if my little Rabbit can keep going.

Caleb:
I mean, the important thing to think about here too is like, right, like all of the ... The majority of the parts are now at a scaled production capacity. The door handles, those are the same as on the Model 3.

Mike:
Right.

Caleb:
Whoever makes those door handles, they're making 6, 7000 of them a week already. Adding another few hundred to a thousand is not going to be that challenging. They could steal some from the Model 3 production line. And so, in the beginning with Model 3, they were going from nothing and getting all these suppliers to believe them and start producing and shipping and get it into their system and have it in the warehouse. They've got these parts in their warehouse already.

Caleb:
The screen is exactly the same. The steering wheel is going to be exactly the same. The mirrors are exactly ... Like thousands of parts are exactly the same. Then it just comes down to can they produce it? Can they actually manufacture it? They can test that because they have all the parts. I do think if there was ever a chance for Tesla to have like higher accuracy of their prediction, this should be it. It's a really big test to see if they have gotten better at this prediction game, especially because they are used to producing 8000 vehicles a week now instead of before, what was it? Two thousand vehicles a week or less.

Mike:
Cool.

Caleb:
Should be good. Okay.

Mike:
All right. What else do we got?

Caleb:
Cyber truck, pickup truck.

Mike:
Oh my god. How did I not hear of ... This is the first time ... Well, I guess about an hour ago, it was the first I've heard of this name.

Caleb:
You started referring to it as Cybertruck.

Mike:
Why? If you hit 88 miles an hour, will it take you back to 1995, like why?

Caleb:
No. You're thinking of the wrong movie. You got to be thinking about Blade Runner, not Back to the Future.

Mike:
No. It's the whole era. Like that was the cyber era, right?

Caleb:
Yeah.

Mike:
Why?

Caleb:
Cyberdyne systems?

Mike:
Yeah. Does it involve time travel?

Caleb:
No, but it is supposed to still look like something that came off of a movie set. Elon has said recently at the most recent like Air Force event he was being interviewed at. He said, it looked really futuristic and this isn't a direct quote, but it would look like you had just dragged off a movie set. People are really going to notice it.

Caleb:
He said on this call, "I think we've said enough about Hessel Cybertruck. We're not going to. This is not the right forum for us to do product launches. But I think it would be, I mean, my opinion, and this could be totally wrong. I could be totally out to lunch here. But I think the Tesla Cybertruck is our best product ever. That's my opinion." Elon Musk. Best product ever. That kind of reminds me of how Tim Cook gets up there and says, this is our best iPhone ever.

Mike:
We're so proud to share this with you.

Caleb:
I mean, yeah. It better be your best product ever. It's your most recent product. He didn't say that about Model Y. He didn't say that's our best product ever. I think they've got a lot of crazy stuff going on in the Cybertruck and I don't think it's going to come out for a little while still and it will be pretty low volume to begin with, but it's going to be bananas.

Caleb:
We've seen some of the leaked image of the roof or the front and I think it's going to look a lot like the Semi where it has this massive raked windshield. I think it's going to be extremely angular. He also said it's bulletproof. I think it's going to have that crazy Tesla glass that the Semi has. I think it might have an option for four motors, one in each wheel, like one for each wheel so it can do like tank mode where like two wheels will rotate forward and two wheels rotate backwards and kind of spin in place.

Caleb:
His brother tweeted out, "I am most excited for the Cybertruck than any Tesla product since Model S. It's going to blow your mind." They're unveiling it November 21st, so 11 days from as we record.

Mike:
The fascinating thing about this, the more I keep hearing about all of these crazy teases and all of this like stuff, the more I'm wondering what the market is for it. Are they going to actually kind of innovate themselves into like the concentric circles of people who are interested and people can afford it and people it's made for? It's going to be like just people named Musk in the middle of it.

Caleb:
That's the risk.

Mike:
Yeah.

Caleb:
I feel like the pickup truck market is unique in a few characteristics. One thing going for it for Tesla is that pickup trucks are high-priced vehicles. It's not unusual to spend 50, 60, $70,000 on a nice pickup truck. There's that. They have the price point to play with. Pickup trucks are extremely high margin for traditional automakers, so they have that price point. However, and then also like towing capacity and utility, I think certainly the electric motor drive train should be excellent for that except for potentially range issues. They have that.

Caleb:
They have the ability to have a voltage, be able to be a generator pretty easily. They're going to have the air suspension so they're going to have a high pressure air for folks who want to have tools that run on compressed air. They can power those. There'll be plenty of capacity for stuff. However, the shape, this Blade Runner-esque shape, I'm not sure it's going to ... It's tricky. I'm not sure how many people who would really want to drive like the Batman tumbler. Is it cool like geeky or is it like aggressive tough? I think that's going to be a pretty fine line because a lot of folks who like trucks like big trucks because they're tough and they look intense and they look strong.

Mike:
Well, yeah. There's been like just a marketing feedback loop going where the trucks just get bigger and more ostentatious like year after year after year.

Caleb:
You just think about how many massive trucks are being never taken off road or really being used for work. They're just really nice play vehicles and just daily drivers for some people. That that's what they like. The same way that SUVs, like big Range Rovers are still pretty capable, but they're mostly lux vehicles.

Caleb:
I'm really, really curious to see if it resonates with traditional truck buyers or of this like brings out people who would have bought like a Model X, but now by this, like what's the seating capacity? Is it going to be five seats or is it going to be like seven? Is it going to have like a covered back in the event that you don't want like a full truck bed?

Mike:
Is it going to be an F150 replacement or like a the Hummer H1 replacement.

Caleb:
Yeah.

Mike:
That's the question is just like, this kind of novelty ridiculous vehicle or is it going to be something that people use?

Caleb:
Yeah. I really don't want it to be like the Aztec, like kind of an in between vehicle.

Mike:
Oh my god. A Tesla Aztec would be awesome. That would be so cool. They should have a have a skin for the Model Y that makes it look like Walter White's Aztec.

Caleb:
Oh boy. Anyways, that'll probably be our next episode. We'll be dissecting the model, the Cybertruck.

Mike:
If we ever get information on it.

Caleb:
Well, November 21st, dude.

Mike:
Okay, all right.

Caleb:
That's only 11 days. I'm going to make the prediction here. I don't think it's that special, but I think it will be towing a rocket.

Mike:
Why not, right?

Caleb:
Yeah.

Mike:
Or maybe it'll be so big, they'll just be a rocket that takes off from its bed.

Caleb:
No. It will tow it, but it won't be an official tow rating that high. It'll have an official tow rating of like 12,000 pounds or something.

Mike:
Maybe they'll tow the Semi in.

Caleb:
Yeah. It could tow the Semi, but that's a ... I think they're going to tow a rocket.

Mike:
What if it towed all of the Tesla vehicles trained together?

Caleb:
Oh, definitely. It could have all the vehicles in a thing and then those are all being towed. I think it's going to do something ridiculous. Because we've seen Model Xs tow like airplanes. It's not safe to do that long term, but it has enough torque to get one of those things moving.

Mike:
You can tow a train like a rail train in.

Caleb:
Yeah. Yeah. I think that will get people's attention. Like it's strong enough to tow a rocket. I'm just so curious at ... Because here's the thing, so many people will see movie vehicles and be like, "I wish that was real." If you really put that on the street next to other vehicles, it would look really weird.

Mike:
Yeah.

Caleb:
It would look really uncomfortable. Remember some of those like, what was that? Yeah, it was like that kind of retro style pickup truck, that Chevy.

Mike:
Oh yeah.

Caleb:
Super bubbly.

Mike:
Yeah.

Caleb:
Then there also is that the Crawler, the Cat Crawler or something? It kind of looked like a 1950s Hot Rod, but it was but built in the '90s by GM. Anyway, some of these retro cars.

Mike:
Yeah. Didn't Dodge make one too that had like weird articulated headlights and everything?

Caleb:
Yes. So I'm very curious but it does seem and sound very Blade Runner-esque. Even so far as the opening credits of Blade Runner reference the setting of Blade Runner and it is November, Los Angeles, 2019. That's why we are unveiling it in November 2019 in Los Angeles. Because Blade Runner is set at that time. This is supposed to be the Blade Runner truck.

Mike:
All right.

Caleb:
Lastly.

Mike:
Lastly.

Caleb:
Our favorite topic.

Mike:
Yes. Do you have a letter grade for this quarter?

Caleb:
Self-driving. We have to talk about self-driving.

Mike:
Oh, well, my favorite topic was letter grade.

Caleb:
Okay. Full self-driving. Over a million Smart Summon usage so far. I've probably contributed about 15 of those. Two or three for like real purposes and the rest for testing.

Mike:
I saw someone use it in a parking garage. I think I was telling you this. I saw someone use it. I was walking into a really crowded Whole Foods and the parking space was so tight. I just saw a guy standing behind his car looking at his phone and just like having a drive in, so he didn't have to like jam the doors to get out, which I thought was so smart parking more than Smart Summon.

Caleb:
See, that's smart, but then how's ... The other people don't have that feature so they're going to have to try and open their door and they're going to ding you.

Mike:
Yeah, that's true.

Caleb:
I don't think it's good to use it to get in because then you're just forcing other people have a really tough time and potentially hit your car.

Mike:
In a more defensive posture.

Caleb:
Yeah. I would take a more defensive posture. I would use it to get out if some other people have parked closely to me, but only after the fact. We got a little detail, that deep scale company that they acquired that had the guy who did I think SqueezeNet. Basically, it was 12 people. It was a traditional acquihire. I think they did about $20 million in Tesla stock as consideration. They had some expertise in increasing the efficiency of neural networks for a given amount of compute, which Elon said was helpful. It remains to be seen. The intent behind that was basically to slightly accelerate full self-driving. That's the intent. Hopefully it will turn out to be true, but no super fundamental technology, really just the people who have expertise on smaller networks for dedicated chips that were running on smartphones. Basically, just how do you tune networks for smaller power consumption.

Caleb:
People were hoping for some massive breakthrough, not really in the cars from that. Then also they're going to be making some more improvements of Smart Summon based on an all the million or so, a subset of those million that have come in. Elon just sort of reiterated that that's part of their advantage is that they have the fleet. They have the data pipelines in a way that they can actually improve based on seeing these edge cases and more unique training data. Nothing super unexpected, just good to hear that there will be some more improvements for Smart Summon. It's definitely early days, but you can see the potential there.

Caleb:
Here's to the full self-driving. That's really was new. He did just sort of unprompted say, "I do. While it's going to be tight, it still does appear that we will be at least in limited in early access release of a feature complete full self-driving feature this year." It's not for sure, but it appears to be on track for at least an early access release for a fully functional, full self-driving by the end of this year.

Mike:
Okay.

Caleb:
Again, I think ... They had said by the end of this year, they'd be feature complete. They're behind but potentially only a couple months behind. The thing that's kind of intriguing about this is what he then went on to say was the next step will be full self-driving with Smart Summon and being kind of the beginning of that. Obviously, we kind of have that two sides of it, highway autopilot and we've got Summon, which is sort of low speed and parking lots.

Caleb:
Now, we need to work on solving the intermediate portion, which is traffic lights and stop signs and navigating through windy roads and windy narrow roads in suburban neighborhoods. That's the focus right now. It's what we've been talking about for months.

Mike:
Yeah.

Caleb:
This is obviously the next major missing piece and how they think about it is not how many of the folks that are going out for full self-driving in a robotaxi world are, where they've been doing like the full experience in very tough environments. Even in certain environments that aren't like San Francisco, they're still way tougher than what Tesla is aiming for because Tesla is really just aiming for like the minimalist features that you really would have to say like, "Oh, the car drove itself." Elon went on further to kind of even further illuminate what he meant. "Feature complete. I mean it's the car is able to drive from one's house to work most likely without interventions."

Caleb:
It will still be supervised, but it will be able to drive. It will fill in the gap from low speed autonomy. Low speed autonomy was Summon. You've got high speed autonomy on the highway and intermediate speed autonomy, which is really just means traffic lights and stop signs. Feature complete means it's most likely able to do that without intervention, without human intervention, but it would still be supervised.

Caleb:
I've gone through this timeline several times, but it's often misconstrued. There's three major levels of autonomy. There's a car being able to be autonomous but requiring to prevent supervision and interventions at times. That's feature complete. I'm going to interject. This is Caleb.

Mike:
See, accents would help.

Caleb:
Intervention at times, that's feature complete. This is where a lot of the media are going to be extremely disappointed and point at Tesla for being incorrectly calling this full self-driving. I think we've been quite clear since this ever was described that this is what Tesla would mean. It's getting closer so now they have to really point out that you are still going to have to be watching and it's still going to make lots of mistakes and they'll still be places where it doesn't work. Just how like Smart Summon doesn't work in all our parking lots. It's like not great in a lot of places.

Mike:
They just have naming problems like autopilot is not autopilot. Full self-driving is not full self-driving.

Caleb:
Yes. It's like it could maybe get from here to the highway, but I still think it might have some problems on some of these turns.

Mike:
It's semiauto, semi-robotic.

Caleb:
Yeah. It's like-

Mike:
It's really more like having a student driver in the car. You've got the brake pedal in the second steering wheel. You might have to take over.

Caleb:
Yes.

Mike:
So, it's really more like a full driver's ed.

Caleb:
But they'll also be like classes of time where it's like it sees ... It's missing something that clearly a human would have seen. I think that's going to be the thing that throws people is-

Mike:
I think ... I mean, I still think that's the most dangerous thing. I mean this is-

Caleb:
Like unknown.

Mike:
Yeah. It's just like people are going to ... If it's just like ... It's going to get good enough, it's going to be good enough where people are going to zone out and people aren't going to be able to recontextualize and take control in time. I think that's going to be the ... The alternative is that it would be ... It would need to know so far in advance that you need to take over. I don't know. It just seems like an insurmountable problem.

Caleb:
The biggest hump I think could be that it's so bad it requires so many interventions in the beginning that there's no way you would disengage mentally. Then that you said training data as people are doing that to basically go from the prototype phase to a really good working version that that might be enough.

Mike:
It seems like the PR on that would be terrible though.

Caleb:
Yes. Yeah. It's going to be rough. I mean, I think it's going to be worse than Navigate on Autopilot. Remember when Navigate on Autopilot first came out? It was really scary to try and use that on the highway.

Mike:
Yeah.

Caleb:
They're like, "Oh yeah, Navigate on Autopilot. Now, you don't have to like change lanes." You still would not want to ... I still didn't want to use in a lot of places. With stop signs and stop lights, like they've been testing that for over a year and a half. The challenge there is like false negatives are so problematic. That's what I want. Yeah, you miss something. That's going to be really risky.

Caleb:
What do you get ... I mean, if you use it, maybe you'll realize like, oh, it normally catches the stop sign, fine. But then one day and where weird light conditions, it doesn't. I mean, they're not using maps. They're not using high def maps, so there is no backup. I don't know. I just really curious. If they're going to be feature complete, that's like they have a month and a half for some people are going to get this.

Caleb:
Even if it slips to January, say we're two months away or two and a half months away that some people are going to get this and there's going to be videos of people using this and holy moly, like this is what I've wanted for a long time, but it's coming and they didn't release it as a safety feature first. I thought they were going to release it as like a warning system first. Maybe they will, maybe they still will. I thought they were going to do it like, oh, there's a stop sign ahead. It won't actually stop, but it will just tell you that it sees it. Then they can use that to make sure that the system's really working.

Mike:
Yeah. It seems like if ever there was an opportunity or a necessity for like kind of a slow iterative rollout like this would be the case. I think they do themselves a disservice by continuing to call it full self-driving and then promise it like a discreet day where it's going to be available, as opposed to something that's just kind of constantly getting better.

Caleb:
I wonder if it's going to do like left turns at intersections or if it's just going to drive through lights.

Mike:
I would treat it like we used to treat first gen Apple products and how I kind of still treat like major Apple software updates. It's like I'm going to let a bunch of other people do it first and then eventually, once I think it's safe, then maybe I might consider doing it.

Caleb:
He then went on and as I said before, at the point where we're able to upload the software, enabling a Tesla to become a robotaxi, expect to have that form of functionality standpoint by the end of next year, in terms of the functionality, basic functionality, aspirationally end of this year, but reliable enough that you do not need to pay attention in our opinion by the end of next year. He thinks it will take just one year to go from the rough feature complete in Tesla's opinion based on the number of miles they have and the quality that you would not need to pay attention. That is wild.

Mike:
We shall see.

Caleb:
I do not think we have seen that much improvement this year.

Mike:
No.

Caleb:
In Navigate on Autopilot. You definitely still have to pay attention. Now, the missing piece is that the current systems aren't really taking advantage of the new 20X better full self-driving computer where this will. But still like holy ... Also, how could they know it's only going to take ... How could they know it's going to take a year now given they don't know what the edge cases are like. That's the whole point.

Mike:
Right. An entirely new type of like software development and all of these unknowns is way more unknown than the rather tractable problem of manufacturing a vehicle and like their estimates are wildly off on those. How can you have any confidence in their estimates of like this crazy thing that no one's done before. It just seems not to slag on them like they should be doing it. I just think they do a terrible job of communicating this process and managing expectations. I don't know how much that is just Elon being Elon, but it seems like it would just behoove them to be a little more conservative and methodical about it. It's still exciting enough on its own like that if you're being conservative and methodical, it's still interesting to talk about and hear about. You don't need to make crazy promises and yeah. If Elon wants free advice, that would be mine.

Caleb:
I agree. I mean, don't get me wrong, I really want to try these features. I want to be able to have the car leave this parking lot and take me to the highway. I don't think that route is particularly challenging. There's a couple stop signs. There's a few stoplights, but it's one left turn and then it's all straight. That should be doable. I still worry about it because there's some areas where there's barely any lane markings and it's a little bit tight and yeah. It's just ... I'm really, really curious.

Caleb:
Now, remember a few months back, they had that autonomy day and they did that loop around 280.

Mike:
Yeah.

Caleb:
That seemed to work relatively well. I think that's basically what we're going to be getting by the end of this year. I would love to have that still, but it's not what these other makers, these other folks are doing in places like Pittsburgh and San Francisco and Chandler. It's nowhere close to that level. People will get it wrong. People will say Tesla is just as good as Cruise and Zoox and Waymo and all these folks. That's just so ... It's just so wrong. There's no way they're anywhere close to that level of quality.

Caleb:
Yet, they're going to kind of try and plant a flag saying they have full self-driving features. We've talked about it, but I think you're pulling ... That's going to be very problematic because they will not be anywhere close to that quality level. It's still going to take years for them to get to that level. That's fine. It's still very fast in absolute terms, but I just wish they were more cautious or they had just ... I don't know how they backed themselves out of this now because they've just painted themselves in a corner where they thought they'd have this by 2018. We're going to be in 2020 and it's still not going to be there. They said they'd have a robotaxis by the end of next year. I want the functionality, but I really feel like they are going to have big, big problems to deal with in mea culpas and hubris discussions again about how far they thought they were going to get with full self-driving.

Caleb:
What we'll hear is, well, there's just so many edge cases. There's so much ... It's just such a tough problem and as you get closer and closer, it gets harder and harder and he's going to make all these-

Mike:
It keeps getting halfway closer.

Caleb:
Yeah. He's going to use that example of halfway closer. You just think and he's going to say, the last 10% is 90% of the work and our team's doing a great job and point all these improvements, but it's going to ... I just want to know, do they realize that today. They just keep sticking to this for aspirational motivation or does he really think there that close? Because he's not dumb guy.

Mike:
Yeah.

Caleb:
He's just not.

Mike:
I don't know. It's vexing.

Caleb:
All right. Well, let's wrap it up. We've got quite a lot of tape here.

Mike:
We went for a while there. Yeah.

Caleb:
Letter grade, you asked me.

Mike:
Letter grade, yeah.

Caleb:
I would say-

Mike:
What's the report card?

Caleb:
This quarter, solid A minus.

Mike:
All right.

Caleb:
Yeah, really saw Ford-

Mike:
It's the profits, isn't it?

Caleb:
It's the profits. It's record deliveries. It's Shanghai Gigafactory. It's Model Y. They're keeping their heads down. They're executing well. No major snafus with the SEC. Then their pipeline is strong with Model Y with the Cybertruck stuff, super excited. Yeah. Definitely continuing to make progress on full self-driving as we've alluded to. Yeah, they got a V10 with the games, with the video stuff, with the Caraoke. Now they have an update going out where it can detect cones, where it gets you 5% more capacity or power, where you've got faster supercharging in the smaller vehicles and also one pedal driving. So that's going out. Yeah, just a really solid quarter.

Mike:
What's been the most popular song in your Caraoke?

Caleb:
Whatever the default was when I tried it. I haven't used it.

Mike:
You don't have to lie. You could tell us.

Caleb:
I haven't used it.

Mike:
You've been singing to Adele, haven't you?

Caleb:
No. I don't. No, I haven't used it. I haven't. I would use it if they started recording from the internal camera because that'd be fun to share out and try and trick people. But they still don't use that internal camera for anything. A huge missed opportunity.

Mike:
Not that they're telling you at least.

Caleb:
I think that would socialize that camera really well.

Mike:
Does that camera have a LED on it? Like does it like kind of like your webcam on your Mac? Like is there a little light that tells you when it's on?

Caleb:
No.

Mike:
No?

Caleb:
That just says you're not using it.

Mike:
Interesting.

Caleb:
Yup.

Mike:
Cool. All right. Well, A minus.

Caleb:
Yes.

Mike:
All right. We get a probably a C minus for our recording regularity, but-

Caleb:
Yeah.

Mike:
It's more than enough.

Caleb:
It's okay. I mean, I-

Mike:
Maybe a D.

Caleb:
Now look, we are still very interested in recording these shows and sharing them. I think our schedules have been particularly challenging.

Mike:
I can guarantee you there will be reaction to this Cybertruck show.

Caleb:
Oh yes.

Mike:
There will definitely be two shows in November. There's no doubt about that.

Caleb:
Yes. That is for sure.

Mike:
Yeah.

Caleb:
You can-

Mike:
The combination of something called the Cybertruck and Thanksgiving break will make for a fertile podcasting season.

Caleb:
It's going to be quite good. All right. Mike, where can people share their thoughts and their opinions about this quarter?

Mike:
Yeah. On Reddit, r/TheTeslaShow. We're most active on Twitter. Our screen name, our call sign if you will is @TheTeslaShow. We also have a website on the worldwide web at www.theteslashow.com. I believe that's the entirety of it.

Caleb:
That's it.

Mike:
I'll talk to you later.

Caleb:
All right. Later, Mike.

Mike:
Bye.

Caleb:
Bye.