I mean, sure. You can make a table that shows production volumes that add up to 2.7 million. It’s easy. Just start by assuming that Tesla will go from producing around ~470K units in Q4 2023 (the Troy Teslike number for Q4) to produce 600K units in Q1 2024, and go from there! Of course, Tesla’ hasn’t done anything like that in their recent past - increase production sequentially by 100K units. But once you assume they’re going to do that this year, the rest is pretty simple.
But I think that’s incredibly unlikely, and I don’t think Tesla will produce 600K units in Q1 2024.
But for a different reason than the OEMs. Tesla’s vehicles were all potentially profitable at their introduction if they could be produced at scale. That’s why Tesla builds its gigafactories in parallel with each car. Tesla’s models were always profitable once mass produced.
VW and GM have both stated they are a couple of years from designing a BEV that can do that. They can each design a decent BEV with specs comparable to the Teslas. But they can’t mass produce them at a cost that would allow profits. The long-term problem is that because they are competing with Tesla, which has a better business model for BEVs, they may never be able to do so.
Tesla designs its production and product around its controlling software. That’s the advantage of vertical integration. It allows optimization of the hardware to fit the software. That’s why Tesla can innovate so quickly. If its engineers find a way to reduce cost by 3%, Tesla can reprogram the robots and reconfigure the parts very quickly. Its why Tesla does over the air upgrades and fixes better than anyone else.
The OEM horizontal business model is to outsource as many parts and software to independent contractors. Their pace of innovation is greatly limited by “what’s out there”. That’s ok with mature technologies like ICEs where innovation is slow. But when it comes to BEVs that are mostly electronic devices controlled by software, that is a losing business model.
The big three solution to that dilemma is to take away from the workers. The big three have been crying that they want their battery plants to be non-union. The union wants the battery plants treated like any other big three plant, ie union shops. The strike was expanded on Friday: Ford Chicago Assembly, and GM Delta Township Assembly.
For the record, I don’t really care if Tesla exactly hits 50% this year or 50%, each year based on a 2020 starting point, etc.
It could even be just 45% or 40% and it would still, over time lead to them dominating the EV car market (at least the US market) compared to the current legacy auto makers that have ~all been stumbling over trying to build EVs, at scale, with profits…and stumbling over how to get them charged (finally fixing that with use of Tesla chargers…but not for another year or two)…and stumbling over their dealer business model that is holding them back (seems only Ford is trying to fix this)…and in the US having to deal with the unions that seem to want a slower EV ramp.
Pretty much any large growth rate, even if it comes in spurts, on average, is going to be a serious problem for legacy auto by the end of the decade.
Note: even a ~40% growth rate gets them to 20M in 2030, if my math is right. But even just 10 -15M would cause major disruptions to legacy auto, even ignoring the additional competition that will come from China.
So do we really care if the number is 40% or 50% and why?
Oh, I don’t think 50% is at all plausible going forward, so I don’t think that’s really what the debate is. Instead, I think there’s a big difference between 20 million units in 2030 - which is what Tesla has indicated is their goal - and (say) 10 million units at that time. Basically a 40% vs. 30% annual growth rate going forward, which might seem a smallish difference but has enormous implications.
A world where Tesla sells 20 million units is one where Tesla is probably the world’s largest automaker, and the largest that’s ever existed - with higher marketshare than any car company in many decades. It’s one where you have single models (like the Y and Future 2) sell multiple millions of units in a single year, which means a remarkable degree of concentration of consumer purchasing. And of course, that’s a world where EV’s are the majority of new cars being sold within the next seven years. That would mean that consumer demand for EV’s stays robust and/or government subsidies and other policies remain solidly behind that transition.
Meanwhile, a world where Tesla sells 10 million units or fewer is one where Tesla is basically the Toyota or VW of 2030 - in both absolute sales and percentage marketshare. It also means that either global EV adoption has been far slower going forward (and the majority of cars being sold are still ICE), or alternatively that Chinese manufacturers have eaten a massive amount of marketshare.
I think Tesla’s growth next year (YoY from 2023) will be closer to 30% than 40%, continuing from where Q4 will probably end up compared to Q4 2022. The delay in getting the CT into production, the saturation of the market for the Model Y, the slower time frame in developing Monterey, and the gradual pulling back of generous subsidies in various markets will all take their toll over the next few years. We’ll see.
But at least from my perspective, the discussion isn’t between 40% and 50% - but whether Tesla can get at all close to either of those numbers, or becomes “just another car company.”
I think you are missing the big picture. Let me bore you with a digression. There was a time when cassette and CD players ruled the portable music business, with Sony the elephant in the room. MP3 players were introduced but initially these were only a niche product. They were expensive and could store about a dozen songs. Then a few years later came the Apple iPod together with better hardware, better software, a ton more storage, and the iTunes store. The iPod soon grabbed over 70% of the MP3 player market, which rapidly made cassettes and CDs obsolete.
Get the drift?
The Tesla models have better hardware, better software, and the best charging network. Those will probably be enough to get them to the Toyota level. To get to historic market share levels will likely require autonomous driving.
If Tesla develops FSD first and better, then 20M annual sales will just be checkpoint on a rapidly rising curve. That’s what that 20M aspiration is ultimately dependent upon. Your mistake is to assume that this is only about cars. It’s not. It is about replacing cars with transportation robots. It is a technological change bigger than going from cassettes to MP3. And that is why it is possible for Tesla to grab 25% of global auto sales.
If that plays out, I wouldn’t be surprised that the gov’t steps in from an anticompetitive stance, makes them break out FSD so that other car makers have equal access before they let the rest of the unionised automakers go out of business.
Nah - I’m aware of developments towards autonomy, and you and I have discussed them frequently. I think you also know that I do not believe that Tesla will be the first or only car company to develop autonomy, nor do I think there’s a “better” when it comes to autonomy. It will be like the software that powers a push-button elevator - it will either work or it won’t, and the user won’t care as long as it works.
Nor would regulators ever allow the scenario you describe to play out, even if Tesla were the first technologically to get to autonomy (arguably they’ve already lost that race, since there are already lots of transportation robots prowling the streets of California). Personally, I think Tesla’s trapped itself a bit with its FSD development strategy. California (to pick a government) is unlikely to let Tesla turn on a million new autonomous cars at once, and Tesla isn’t set up to have a pilot program where only a handful get turned on for a while. But that’s the low bar. Whatever happens in the U.S., the governments in China and the EU simply are not going to let Tesla turn on FSD in their jurisdictions if it means Tesla will have years and years of destroying the domestic auto industries there.
As for your specific example, cars aren’t iPods. Apple was able to go from selling 400K units in 2002 to 8.2 million units in 2004 to 40 million units in 2006. That’s because they didn’t have to build their own factories to make their own product - they were able to farm production out to an ecosystem of existing consumer electronics manufacturing firms (like FoxConn) throughout the world. You just can’t do that with automobiles, where it takes you three years (at least) for a new factory to go from permitting to ramp up to full production. So even if Tesla had a few years of “first and only,” it could only increase volumes by so much before that period ended and autonomy started to become more ubiquitous.
You seem to be talking like 2030 is some kind of end point. If Tesla does (some day) reach 20M units it doesn’t have to be by 2030…even if they have said this is the target date. I doubt they can actually do it by then. But then, they seem to be the only one projecting large growth (in EVs) that is actually growing significantly (China excepted).
Don’t we already know that they aren’t just another car company?.. given that they are the biggest player in fast chargers and doing batteries for home and grid. Batteries are really important since it allows them to potentially develop economies of scale in battery materials that legacy auto doesn’t have. They can develop/expand multiple battery technologies and shift supplies to/from autos as required.
I mention it just because that’s Tesla’s stated goal. It helps put the growth rate in context relative to their projections. I obviously don’t think Tesla’s getting to 20 million units by 2031 or 2032 either.
Again, this is part of why I find discussing EV adoption fascinating. Even if the above statement were true (and I don’t think it is), why “except” out China? China’s what matters in this discussion. About 60% of the world’s EV sales take place in China, and growing. The Chinese market is bigger than the rest of the world combined, and it’s growing much faster than the rest of the world combined. A big part of the reason why Tesla won’t grow to 20 million units in the next decade or so is because you can’t sell 20 million units unless you’re dominating (not just leading) in China, and Tesla won’t be dominating in China. They’re not dominating now (less than 10% of EV’s and less than 15% of BEV’s), and now’s as good a competitive playing field as China’s ever going to let Tesla have against their domestic manufacturers.
I mean - not really? Sure, they do those things (and solar, too) but they’re a tiny, tiny part of the company right now. Less than 4% of revenue, IIRC?
But I don’t want to sidetrack the thread into a debate over whether Tesla’s powerwall business might one day be a more meaningful part of its revenue stream. I was talking about Tesla’s automotive business - like we might talk about Nissan’s or Mitsubishi’s auto segment, even though both companies do a lot of other things.
If Tesla can grow to 20 million units (and therefore >20% marketshare) as fast as they project they will, they will be unlike any other automobile manufacturer in history. If they “only” get to 8-10 million units…that’s outstanding, but not revolutionary.
I do think that the powerwall and grid batteries (mostly grid batteries) are important for Tesla. Tesla can make bigger investments on materials for batteries and have the risk minimized (compared to legacy auto) BECAUSE of the other businesses. Years ahead planning is needed to have enough batteries to grow the EV business. If they have excess kwh of battery materials they can funnel this to the grid. Likewise, if they have a shortage of materials they can delay some grid projects and/or slow down powerwall shipments (and/or do price adjustments).
Yes, the different batteries don’t use the same cells. But ramping up big battery volumes requires long lead times to secure mining contracts, etc. We’ve already seen big plans by legacy auto on battery factories, then later delayed. No factory, no need for materials means more potentially available for Tesla.
Again, I agree that getting to 20M is a long shot, especially by ~2030.
But then if the don’t plan for the batteries to supply them they won’t even get close.
I guess maybe? Their energy storage business is really small compared to their automotive business. They had about 6.5K Mwh in battery energy storage systems last year, compared to just about 95K Mwh worth of batteries sold in vehicles (figure just over 70kwh per car, since most of them are Y’s with a few X/S thrown in). Not sure that’s enough to really matter, particularly given the fact that Tesla is pretty insistent that it’s got so much demand for the vehicles.
It might be a hedge against regulatory changes - if auto subsidies get phased out for various reasons, you might see those funds shifted to electricity generation/storage subsidies. And of course, government-owned or heavily regulated institutions like institutions might be willing to pay premia for green energy going forward. Still, this doesn’t seem like something that’s big enough to fundamentally change what kind of company Tesla is.
But as you point out, the batteries aren’t interchangeable. Building a Powerwall factory (like the new one in Lathrop) doesn’t provide extra capacity to produce batteries for their cars. Just Powerwalls. If you need X number of batteries in year Y, it doesn’t help you to have Powerwall capacity in a factory that can’t produce car batteries. Sure, it provides a nice “sink” for you to have big contracts on extra lithium - but then again, why do you need one? If long-term lithium contracts are an important supply bottleneck, why aren’t you using that lithium to build more car batteries sooner?
I mean, one answer is that you foresee the possibility of a long-ish slowdown in growth of EV adoption at some point in the future, with consumer demand cooling off materially. But honestly, that answer is probably more consistent with my view of the future market than the contrary - that Tesla’s automotive side can’t grow as fast and as high as they’ve been talking up.
Maybe I didn’t spell it out. The hedge isn’t on shifting batteries from car to grid batteries or grid to car batteries. But on establishing favorable pricing in the long term contracts or joint ventures in the mining of raw materials and then being able to shift supplies, as required.
Pretty much the same as Tesla’s unit sales. From 2018 (the earliest I found) they’ve grown their battery installations by about 58% annually. Their vehicle sales, meanwhile, have grown about 52% annually over that time period. Since the model mix shifted from nearly all Model 3 to mostly Model Y, I suspect that the proportion of batteries in BESS to vehicles has barely changed.
No, I get it. It’s just that with such a small amount of supplies going into BESS relative to vehicles, is that really a material factor? Tesla’s already one of the biggest purchasers of materials for these types of batteries in the world - is adding an extra 6% onto volumes likely to be significant? Or is this more likely the other way around - Tesla’s getting favorable pricing on battery materials already, and this is an efficient way to leverage that into another business line?
In the first quarter of 2023, Tesla’s energy business accounted for 6.6% of the company’s total revenue. This is a solid increase from the recent quarters because of the combination of the energy business growing revenue a whopping 148% year over year and the auto business growing revenue “only” 24%.
That’s good news for the energy division, for sure! If that keeps up, then they will certainly become a more significant part of Tesla’s operations.
No, it’s the reverse. If I get a discount of 10% because I’m buying 100 units of lithium for my auto business, and they let me up my order to 106 units at the same rate, then the extra 6% that I’m adding on isn’t improving my cost to the auto segment. It’s me taking advantage of the discount that I would get regardless of whether I had a BESS segment, because my auto segment is so huge. It’s certainly a way I can leverage that discount, since it will give me an advantage in that other business segment.
That’s the underlying idea behind lots of conglomerates, and especially the heavy industry conglomerates that some automakers are part of. If you’ve already developed a lot of resources in making cars (access to large supplies of rubber and steel, expertise in heavy manufacturing, etc.) you can leverage that out into making other things: construction equipment, marine and shipbuilding, power generating equipment, machinery, etc.
The pros to that approach are the ones you note - pooling worker talent and access to supplies across multiple divisions can generate efficiencies and potential cost savings. But with those benefits come the disadvantages of conglomerates - these divisions are in some ways rivals with each other for resources and talent within the company, and having multiple businesses can cause a company to lose focus or fail to properly execute on one or more of its businesses.
I think that is the crux of the matter. The Tesla bull narrative is that BEVs and self-driving are going to be revolutionary with Tesla currently holding substantial tech advantages in both. You assume that BEVs are just another type of car and will follow the same business models successful with ICEs.
On another note, the most mature BEV market is Norway where plugin electric registration in 2023 so far makes up 90% of total car sales and where there are a ton of brands and models competing. The Model Y makes up 20% of all cars sold so far this year, with that percentage rising year-to-year. That’s all cars, not just electrics. The Model Y grabbed 24% of total car sales in September. I’m guessing that percentage could increase if Tesla was also selling a $25K model.
Note that total car sales were down almost 30% in Norway due to a new VAT on electrics and interest rate hikes. Tesla sales has demonstrated to be far more resilient in a generally declining market than its many competitors. That’s evidence of a big demand advantage.
That’s a good way of putting it. For BEV’s, I think it’s going to be more like the transition of televisions from cathode-ray tubes to LCD’s - a significant incremental change, and one that has a number of impacts on the manufacture and use of televisions, but not revolutionary.
Self-driving vehicles would be more on the revolutionary side. I think that tech represents far more than an incremental improvement. There, we disagree on the relative position of Tesla’s efforts. I’m just skeptical about whether Tesla’s strategy will actually result in a substantial advantage for them.
They’ve done amazing things in trying to solve the broadest type of self-driving problem - Level 5 autonomy using almost entirely vision. But I think there’s a very good chance that they won’t be able to solve that problem until well after many other auto companies have fully and thoroughly solved Level 4 autonomy using other sensors and tools - at least for enough areas that solving Level 5 isn’t really that much of an improvement. You don’t need to solve the general case if you’ve solved all the specific cases that really matter. And since Level 4 probably works better with existing regulatory and safety frameworks, at least during the initial adoption phase, Tesla’s adding to their difficulty even beyond the greater hurdles in developing the tech.
Maybe it depends on how you define television. If you are thinking of studios making shows and movies and sending them (somehow) to your home to watch, then, going from CRT to flat was not really revolutionary. But if you include all display monitors, then I’d argue the opposite.
When we had only CRTs about the only place you saw one not as an in-home TV was showing flight status at an airport or one or two at a sports bar.
But now we see them showing the menus at restaurants, phones in our pockets, on laptops, on our wrists, on signs up and down the Vegas strip on photo frames in homes AND on our walls at home so we can watch reruns of all those TV shows from back when all we had was CRTs.
Live concerts are flooded with flat screen displays.
TV/movie production have begun using what is called virtual production where the old green screens are replaced with giant flat panel pixel walls that use game engines to change the background’s 3D perspective as the camera pans and scans to new positions. (Yeah, some of these aren’t actually LCD, but LEDs…but not even close to possible with CRTs)
But revolutionary in their own ways.
