There’s another dimension to this, which is the the “contribution margin” per kWh.
As long as Tesla is battery cell constrained (e.g. next 5 years or longer) every kWh of cells they sell needs to be attached to a certain amount of profit. This is why Semi is not going to ship in volume for a while.
There’s rumor Tesla will build a cheaper compact car for China and EU. If you allocate 45kWh of cells to a car which sells for $20k and earns $4k of profit, versus 75kWh of cells to a car that sells for $50k and $10k of profit, all else being equal (20% GM) you want to sell more of Option B because the battery cell is your manufacturing limit and profit per kWh is $133 vs $89.
Or to put it another way, a car that clears $4k for Tesla can only have a 30kWh battery, or else selling that car drives down profit.
Tesla is rumored to be planning on making their own chemistry cells at GF4. I expect we will see real volume in their next-gen chemistry by the end of 2022 or mid-2023, at which point their annual production run rate will likely be greater than 1.5m vehicles requiring 100GWh of batteries per year. Currently they are closer to 35GWh, and IMO scaling the battery production is by far the hardest part.
They have started using LG Chem cells in China, whereas up until now they were only sourcing cells from Panasonic. This is because it’s simply not feasible to produce the needed cells for GF3’s 150k/year run rate at GF1 and ship them overseas. Although that’s what they did for the initial China production run.
They'll sell the most lucrative products first while ramping up the R&D to tap into more markets when those eventually saturate. So, 30kwh battery car doesn't make any sense for tesla because they'd eat into the sales of cars that make much more profit per unit. The model 3 has shown that they are really good at upselling software features and extra kwh (read even higher profit margins).
Their strategy seems pretty clear to me as they have plenty of products in a late stage of R&D that they could sell many off if only they had the production capacity to meet demand. They are basically tackling this in order of profitability. Model Y is next. They have the Tesla semi ready to go as well in limited volumes and they've been talking about a new roadster and model s as well. That should keep them absorbing whatever amount of kwh they can squeeze out of their factories for years to come.
So, I don't see what value a low capacity cheap car would add. They'd need to sell many more of those than of their other products to make the same amount of profit and they are basically bottle necked on production capacity already. It doesn't make sense for them to even consider sacrificing factory space for this until after they build a lot more production capacity and the market for higher end options is saturated (which will take quite a few years).
They haven't even doubled R&D with the Model 3, from a low base, and R&D has actually DECREASED since 3 years ago. Where do you see them "ramping up the R&D"?
Tesla is averaging about $1.5B annually in R&D spend. For comparison, GM spent $7.8B last year. VW spent over $13B.
Tesla aren't even investing fast enough to keep up with their capital depreciation, but somehow people keep saying things like, "they're re-investing all their profits". If they are, show the data.
>They have the Tesla semi ready to go as well in limited volumes
GM has revenue of around USD 150B. VW of around USD 260. Tesla's was USD 20B in 2018. So compared to revenue, Tesla spends much more on R&D. And don't forget that a lot of the traditional car manufacturers' R&D is spent on combustion engine improvements and other tech that is not related to EVs whereas Tesla is only spent developing EV tech. Further, Tesla has been developing EV tech for 17 years now, the last 8 years at some scale. The traditional car manufacturers have only just begun in earnest 3-5 years ago.
I do agree that Tesla isn't investing that heavily in R&D but they also cannot afford to. They are not turning profits. And they are already 5-10 years ahead of many competitors on tech, so I think it's wise to invest in production capacity instead. This year and in 2021, VW, GM, Mercedes, BMW etc. are getting into the EV market with cars that in specs seem to be competing with Tesla's 2012 model S.
The most important spec right now is price and range. None of the old companies are close to delivering competing range at a competing price. None of them have a charging network or battery efficiency or battery production capacity close to what Tesla has.
>So compared to revenue, Tesla spends much more on R&D
No, they don't spend "much more".
Old, dying GM and high-flying, take-over-the-world Tesla spend nearly identical amounts as a percentage of revenue. How can that be possible? One is building factories and semi-trucks and solar roofs and AI chips and robo-taxis and "ramping up investment" and the other is going out of business (so I hear).
>None of them have a charging network or battery efficiency or battery production capacity close to what Tesla has.
Tesla arguably makes the best EV on the market right now, that's true. But this idea that no one wants anything else, once again, flies in the face of the data. Competitors are selling well all over the world. And that's what we want, isn't it?
>None of them have a charging network
The charging network was a genius way to get people to buy EVs at time when range anxiety was an issue. It was one of Elon's best moves.
But the days of it being a real competitive advantage are almost gone. Car companies don't own gas stations. It's cost center. Did you see the lines over the Thanksgiving and Christmas holidays? Who wants that?
R&D isn't the only place you can invest profits. Building some of the world's largest factories, filling them with equipment, and hiring employees to fill them, is not cheap upfront. It is an investment which should pay dividends.
Semi is likely at least a year away from any significant scaling.
Show us where it shows up on the financial statements then? I bet you cant. Because after stiffing Nevada[1], Buffalo[2], and California [3], Tesla went hat-in-hand to China for special loans and promises to pay taxes there[4]. I guess you could say it's "smart business", in the same way that PE buying a company, gutting it and selling it off for parts is "smart business". But make no mistake: China owns that factory.
This is not conspiracy, it's in their disclosures. I simply cannot understand what people root for a company doing these things.
I don't think Tesla is fundamentally cell constrained on that timescale. They can expand cell production a lot faster than that. They have plenty of site space at Gigafactory 1, so can probably go from decision to expanded cell production inside of 18 or 24 months (maybe less if they find an empty factory with suitable utilities). Even faster, they can buy cells from, as you say, LGChem or others.
Tesla has long bought cells from others for its stationary storage products. They could pursue a similar strategy for Semi if they wanted (in the near term). The lower range Semi does not push battery chemistry particularly hard, so a third party cell is easily feasible there.
So I don't buy that they're cell constrained on such a long timescale. They're constrained by other things.
Will there be an additional 65 GWh of suitable battery cells available from third-party suppliers globally, at reasonable cost, in such a timeframe?
This seems somewhat doubtful to me, given how strategically important battery cells are for EVs, and how everyone and his dog has announced that they will make an EV.
Exactly 65GWh? I dunno. But a substantial number, yes.
Announcement of EVs doesn't count for much. The last 5 years have seen lots of announcements and not a ton of deployment.
Cylindrical cells (except for the longest range or highest performance vehicles) are nearly commodities. Tesla's early decision to go with cylindrical cells keeps being validated. It's a major risk reduction to be able to switch cell producers without redesigning the whole battery pack or car. Other car companies like Rivian are also using cylindrical cells.
"Tesla is rumored to be planning on making their own chemistry cells at GF4. I expect we will see real volume in their next-gen chemistry by the end of 2022 or mid-2023, at which point their annual production run rate will likely be greater than 1.5m vehicles requiring 100GWh of batteries per year. Currently they are closer to 35GWh, and IMO scaling the battery production is by far the hardest part."
In addition Musk seems to be saying they will need an incredible ramp in battery production (to 2-3twh/year!!!) to make an important contribution to global warming. Given his style, I wouldn’t doubt he intends to have Tesla do the manufacturing: Q2 2019 earnings call:
I could see them supplying cells for stationary energy storage if Tesla energy is at capacity, but I suspect they would only become a cell supplier to another auto manufacturer if that manufacturer was serious about EVs and willing to commit to a long term contract.
Sure, but batteries are far more 'core' to Tesla's businesses than trucking or car carriers. TBH, I never understood that comment by him. I imagine he got interviewed right after some Tesla operations meeting where they were struggling to find trucking capacity and he just blurted out 'lets build car carriers' to the nearest microphone.
>Sure, but batteries are far more 'core' to Tesla's businesses than trucking or car carriers.
That's true, but I'm not sure it makes any more sense than car carriers. The cells that Tesla users for their battery packs are commodity products. Why would you bother to try to manufacture those? I mean, batteries are core to Apple's business too, but they aren't producing them.
The value-add is in their pack technology. The move back to vertical integration is odd.
>The cells that Tesla users for their battery packs are commodity products
Not in the volumes Tesla will require over the next decade. I have a hunch that every EV manufacturer that tries to get serious about volume production (> 1 million vehicles per year, say) will hit a brick wall of supply shortage. There either won't be enough cell production capacity available, or competing manufacturers will bid it up enough to seriously impact the profitability of the vehicles.
Until EVs are established as the obviously best alternative for most vehicles, global production capacity will lag to the degree that in-house battery cell production is a necessary strategic advantage.
Add that to the fact that the ~90% efficiency of electric drivetrains makes single-digit improvements in component properties hugely impactful on the economics of the final product, and battery cells seems like something that should mainly not be outsourced.
If the cells are truly a commodity, the only thing in-sourcing gets you in terms of profit is the opportunity to beat the market on demand prediction (that is, you may over-spend on production and not actually benefit from your capacity predictions).
There's a similar effect with the component properties. Really, you are looking at the properties per dollar spent, so if you take a wrong term, in-sourcing costs you relative to the market.
Tesla have shown a rather large appetite for vertical integration. For example, they make their own seats. Why? I guess they think they can eek out some advantage. They could certainly get commodity seats from a traditional auto supplier, but they must have some line-of-sight to making them cheaper, or better or maybe just more customized than what a supplier would produce.
I think the same will be true for batteries. They won't be making a copy of the Panasonic cell design. They will only become a manufacturer if there is some advantage to be gained. They will probably seek to integrate IP from their Maxwell acquisition and improvements from their research group in Canada.
>They could certainly get commodity seats from a traditional auto supplier, but they must have some line-of-sight to making them cheaper, or better or maybe just more customized than what a supplier would produce.
That's what they think. What do you think? Do you think it's possible for a single player to be the best builder in everything?
When it comes to manufacturing, the world is flattening out. You can produce things from components built and sourced all over the world. But Tesla is going in the other direction. One has to at least question their reasoning, and not assume "Elon knows best".
Personally I’m rooting for Tesla’s vertical integration approach. Tesla doesn’t need to be the best, just good enough to make a component of adequate quality for less than what the supplier would charge for it. I do think they’ve figured out some places where they can get to that level.
For commodities like semiconductors it would be insane to try and open your own fab: those players are so much more sophisticated than anyone else. For mechanical components I think they logic is a lot more complicated. In a phrase: it depends. I’ve been working in product development for a few years and have been to overseas factories to help set up assembly lines. Some suppliers are fantastic and we won’t beat the quality / price. Others just don’t seem to get it and we’ll spend months and many engineering hours basically teaching them how to build what we need. It’s those cases where I would personally prefer to be more vertically integrated.
> They also bought a trucking company and said they were going to start building their own car carriers
Surely once they actually have full self driving nailed the cars will just deliver themselves. It's just a question of where in the 1-20 years away that is.
Another take is that Panasonic’s GF1 venture isn’t very profitable, so they’re unwilling to do more with Tesla. They may expand production, but only at a moderate pace and only after seeing sustained demand. That’s why Tesla is abandoning them for China.
The only thing I can think of that could eventually offset profit per kWh is if their ADAS offerings get significantly better with a larger and more varied fleet.
In that case, trading lower profit per kWh on the initial sale for more cars/situations to train their models might be worthwhile, especially if better ADAS also encourages more people to purchase it when they buy the car or after.
I agree with your analysis. Let me just add that Tesla keeping its profits high means it has more money to build new car factories and also expand battery production.
When picking gross margins it is not always desirable to pick the highest. It would be unusual, but not actually illegal or wrong to pick a mix of gross margins. It depends what your goals are. Does anyone think VW only makes and sells it's highest gross margin products? Or Heinz or anyone else?
It's an interesting analysis but it doesn't define what tesla will do or why they do it. It only defines what sets the upper and lower bounds on expectations of return on investment.
VW sells a big mix because it already has built all the factories needed to do this. Tesla is still expanding rapidly, so makes sense to focus on the most profitable products for now.
As long as Tesla is battery cell constrained (e.g. next 5 years or longer) every kWh of cells they sell needs to be attached to a certain amount of profit. This is why Semi is not going to ship in volume for a while.
There’s rumor Tesla will build a cheaper compact car for China and EU. If you allocate 45kWh of cells to a car which sells for $20k and earns $4k of profit, versus 75kWh of cells to a car that sells for $50k and $10k of profit, all else being equal (20% GM) you want to sell more of Option B because the battery cell is your manufacturing limit and profit per kWh is $133 vs $89.
Or to put it another way, a car that clears $4k for Tesla can only have a 30kWh battery, or else selling that car drives down profit.
Tesla is rumored to be planning on making their own chemistry cells at GF4. I expect we will see real volume in their next-gen chemistry by the end of 2022 or mid-2023, at which point their annual production run rate will likely be greater than 1.5m vehicles requiring 100GWh of batteries per year. Currently they are closer to 35GWh, and IMO scaling the battery production is by far the hardest part.
They have started using LG Chem cells in China, whereas up until now they were only sourcing cells from Panasonic. This is because it’s simply not feasible to produce the needed cells for GF3’s 150k/year run rate at GF1 and ship them overseas. Although that’s what they did for the initial China production run.