The old-line ICE automakers want to make a pathway for EVs. However, they are constrained by a variety of market forces–workers (both of the companies AND their suppliers) PLUS the many millions of people who are unwilling/unable to quickly “jump” into a totally new technology from known mfrs (old-line auto makers) making their first few models of these new types of vehicles. The cost of the battery is the #1 problem–but the likely solution is not yet here. Costs for batteries keep dropping as the available power and faster recharge rates become available. Hopefully, 3-4 years? Meantime, whatever gets built NOW is not going to get the battery the mfrs (and customers) want. So it makes sense for projects that take 2-3 yrs to get completed begin now, but be able to readily pivot to the new tech when it becomes reasonably available. Hurry up and wait…?
My guess is that proliferation happens when providers have nothing significantly newer and better to offer so they just add more kinks to the metal, chrome, tail fins, and other minor ‘improvements’ to differentiate their offerings from dozens of look alikes. Tesla can get away with four models only util competitors come up with similar products which is why Tesla needs to innovate at warp speed.
And Tesla is planning on doubling the number of models they offer in the next few years. And lots of other companies are adding numerous new EV models all the time - as well as brand new companies (like Rivian) coming on line with their offerings. And consumers in the western world are about to get exposed to literally scores of new Chinese models in the coming years. There might be half as many companies as there are now - but all of them will be new to those western markets.
You can’t just look at the trend lines that go down, and ignore the trend lines that go up. The major auto markets are always frothy with models being discontinued, and new models being added. German buyers may not have as many VW, Mercedes, or BMW models to choose from in ten years - but they’ll have a host of new BYD and SAIC and GAC Alon models to pick from.
Oh, and consolidation in the industry doesn’t mean that the models have to go away. Chrysler got consolidated into Fiat - and then they both got consolidated into Stellantis. But you can still buy a Ram pickup or a Jeep Wrangler or a Chrysler Pacifica. Models, and even the brands, still persist after a corporate merger.
Again, cars aren’t like cell phones or computers. Their main function isn’t electronic. It’s to move people and cargo in the physical, real world. Which means that their physical form directly affects how they can perform that function in a way that isn’t true of a phone. Consumers have different needs in terms of what their cars have to do for them - trade-offs between passenger space and cargo area, overall vehicle size and cost, specific form factors (like pick-ups), etc. And they have different preferences that can only be met by different models - you can’t make a car into a convertible with an OTA update.
Yes. I understand that Ford is projecting profitability in this segment in 2-3 years, while it took Tesla 18. Why is it so surprising to you that a new segment isn’t profitable on Day 1?
Because there are already other companies and countries doing it? And batteries are a developing commodity product, and there are already dozens of such factories around the globe. The US is already #2 in world production, projects doubling within two years, and raw materials costs have dropped 50%. This is not really complicated.
Probably not the ones that are already profitable and delivering more cars in different segments than Tesla, tho. The US auto industry had hundreds of manufacturers in the beginning. Did all of them disappear? No? Quell surprise.
However, most of the major OEMs are not waiting. They are instead trying to copy the Tesla model. GM and VW are trying to vertically integrate battery production with their own versions of gigafactories. Most are shifting to a small number of vehicle platforms in order to make production more efficient. They are just several years behind Tesla. That is a long time.
That is the paradigm shift that is happening in the auto industry. To change horsepower or fuel efficiency in an ICE, different engines are needed. One goes from 6 cylinders to 4. Changing handling and driving response requires analogous hardware modifications. That’s why there are so many ICE models. You need to make a different vehicle to have different performance. BEVs are radically different. Tesla has demonstrated that one can substantially alter the performance of the same vehicle by simply changing the software. The need for different model platforms to provide different characteristics is much reduce.
Autonomous driving will magnify this paradigm shift. Even at levels 2-4, well below complete self-driving, consumers will use these systems to do less actual driving. The less active driving by humans, the less important stuff like acceleration and handling become when attracting customers. More important will be stuff like comfort and interior space.
Spacious boxes-on-wheels become much more attractive than the cramp confines of a sports car if you aren’t doing much of the driving. Particularly with a population that is growing increasingly obese.
While this is true, it is true of any business. In addition to “getting away with”, I think Tesla consciously chose to have fewer models so they can focus on profitability and volume instead of on models. I strongly suspect that had they pushed the production of the cybertruck and/or the roadster in 21/22, they would have sold far fewer 3 and Y models in total, AND would have sold far fewer vehicles in total, AND would have earned much less gross margin in total.
Again - is it? Is the driving performance why we have compact cars and midsize sedans and full-size sedans and convertibles and SUV’s and cross-over SUV’s and pick-up trucks and passenger vans and hot hatches and cars with unique styling (like the PT cruiser or the VW Beetle)?
Well, sports cars are only a tiny tiny part of the market. Virtually all cars today are already pretty spacious - about 75% of all vehicles sold in the US are SUV’s and light trucks. Small and midsize cars are barely 15% of the market:
I also suspect that for a very long time after autonomy starts getting implemented, most sports cars will be driven by humans - just like a good chunk of them still have manual transmissions instead of automatic ones. It won’t be for many decades, if we decide to prohibit human driving, that that will really change.
How many of them still have manual transmissions? It’s difficult to find any cars with a manual transmission anymore. Outside of “supercars”, there are very few.
Only about 1.7% of cars in the U.S. have manual transmissions. That’s also roughly the market share for “sports cars” (which isn’t a very rigorously defined category). I’m sure the overlap isn’t perfect, but I would expect that most manual transmissions are chosen in the sports car segment than any other.
That is certainly a possibility. It is hard to know exactly how and why they came to their choices. Without profitability growth becomes much harder and going to capital markets dilutes shareholders.
No, it’s not hard to know, since Tesla has been quite explicit in why it has made the choices it has. Production has been limited by battery supply. They said that if they had added new models before now, that all they would have accomplished is to build fewer of one model in order to build more of another. So really the only thing that made sense was to do the most efficient production possible, which meant growing their existing mass production car production as quickly as possible.
They have said recently (in multiple quarterly conference calls) that now is the first time that they have all the batteries they need, hence Semi (which has already started production) and Cybertruck, and more next year. Also, they are finally ramping up energy storage in the form of Megapacks and Powerwalls.
Makes sense. On the other hand, and not being a car expert on the level of Mona Lisa Vito, I can’t help but wonder how many car manufacturers start life with more than one model. I know Ford didn’t. I know that GM didn’t spring to life with the ladder already in place; it was a consolidation of individual manufacturers each with a single model.
When Westinghouse provided managers car they had to be American, except we got word that they were changing that policy and if we waited a few months we could choose - up to a dollar range. So I waited. Right about then the Japanese launched their upscale lines: Acura, Infiniti, Lexus. Acura and Infiniti offered two models, one new and one a rebadged Honda that had been sold in Japan already for two years. Same with Infiniti, one new, one a rebadged Toyota. I don’t know about Lexus, it was above the cut-off point so I never looked. But my choice was “color”, not multiple brand new models, there was just one new and one “old” (but new to the US) from which to choose.
Maybe when you’re starting out a highly capital intensive business with multiple production steps it just makes sense to start simple and branch out from there?
We Tesla cultists have argued in defending Tesla’s initial gigafactory strategy that the most significant factor distinguishing BEVs is the battery. This was demonstrated in 2022 where the availability and cost of batteries determined the production numbers, price, and profit margins of the various BEVs on the market. Not surprising that Tesla did well when others floundered.
In the long run though it will be the software that matters the most. Goldman Sachs projects that in a few years, vehicles will required coding far more complex than that currently found in cell phones and fighter jets. Software Is Taking Over the Auto Industry
This I believe will drive the reduction in car models as they will all be derived from a small number of standardized platforms. The distinction between models and makes will be less on how they look physically and more by how they perform based on their software.
Who will be the non-Chinese winners in this future (I phrase it this way because I find China unpredictable)? We have a number of car makers with production expertise. We have a few tech companies that are really good with software and are dabbling in car operating systems (Apple, Alphabet). And we have Tesla that has long focused on both.
That is what all manufacturers of all products would like. It makes their job easy. Unfortunately for them, consumers have other ideas, which is why you find such a plethora of makes, models, styles, colors, and feature options of every product imaginable, from microwave ovens to washing machines, from cell phones to personal computers.
Here is a partial list of those working on self-driving softwares:
Surely not all of those will be successful, but there will be far more than one. This is unlike cell phones or laptops or PCs, you don’t need a “network effect” here. Each software will control a single unit, it does not need to interface with others to form nodes or have brains. And, as we have seen, it is fairly trivial to write code for a smart phone once you know what the target is. Apple did it, of course. So did Google with Android. Then there was Microsoft, but it lacked the network effect because it came too late, not a fault of the software. Likewise let’s not ignore Harmony OS or the Linux based COS popular in China, or BharOS from India. Only two are huge for now: they got their firstest, fastest, and the network effect is real. For automotive it doesn’t have to be.
Final note: there are more battery factories operating this year than last, last year than the year before that, and likewise back for a decade. Next year there will be more, and more after that. Batteries will be branded, but with possible well marketed exceptions, generic. Nobody is going to be buying their car based on the batteries inside, just as almost nobody buys gas because of the big sign on the street advertising “Texaco” or “Marathon” any more.
This is going to be a fun race, with winners and losers, and I am quite sure that in a decade or two the market will look vastly different from the oh-so-confident predictions of today.
Your post, taken as a whole, makes sense but it has a conceptual flaw, software can’t store energy just like brains can’t. It’s like arguing that brains created the Industrial Revolution or that Turing’s Bombe won the Second World War. There is a song that tells it well, “Love and Marriage, Horse and Carriage, you can’t have one without the other.”
Enigma and the Bombe
Turing played a key role in this, inventing – along with fellow code-breaker Gordon Welchman – a machine known as the Bombe. This device helped to significantly reduce the work of the code-breakers. From mid-1940, German Air Force signals were being read at Bletchley and the intelligence gained from them was helping the war effort.
Intellectual and physical need to be kept separate, no amount of brains can reduce the density of lead but brains can find the substitute for lead just as brains found substitutes for brute power, for sails, for steam, for whale oil. Keeping the intellectual and the physical separate leads to better understanding.
Human created code is being substituted by AI created code, possibly the greatest leap in the intellectual domain to date. Quantum brains vs, biological brains… Humans obsoleting humans…
Again, I think there’s absolutely no reason to think that’s true.
All of the main functions of phones and computers are electronic. So people care almost entirely about their software, not their physical form.
Nearly all of the main functions of cars are not electronic. The primary functions of a car are to move people and cargo from one place to another in the real world. Which means the form factor of the car is, and will always be, vastly more important to performing its functions than the form factor of computers or phones.
In a fully autonomous auto market, I don’t think more than a handful of consumers will care about their car’s software at all - any more than they care about the software powering the elevators they ride in a building. They will just work. No car will be allowed to be autonomous unless its software meets a very high minimum standard of performance - so high that differences between auto companies will be relatively trivial.
But - there will always be a world of difference between a pick-up truck and a two-door coupe. Between an SUV that seats 9 and a VW Beetle. Between a passenger van and a convertible. And the physical form of the vehicle - passenger space, cargo space, styling, fit and finish, interior appointments, even the number and placement of the GD cupholders - will still be the most important differentiating factors for consumers.
Just a side comment here. There was more than one cause for the Microsoft mobile effort failure, but it DEFINITELY included software. At the time, most phones were acquired via the carriers, and the carriers simply could not get the Microsoft phones to work well with their systems. Even me, a mobile phone developer for many years at the time, along with my carrier(s) highest level support teams, couldn’t get the Microsoft phones to work correctly. For my own family, I purchased 5 of them via the carrier, they could get certain features to work, but not others. Then when tinkering to make features work, other things would break. Early on, once in a while, a phone would brick, and they would replace it. There were software upgrades all the time. Finally the carrier agreed to replace all 5 units with another, higher end, Microsoft product. Then spent another few months trying to get those to work properly. No luck. Finally we all gave up and they sent us 5 new phones from a different manufacturer (LG) that worked fine. And this wasn’t just “a few hours” with tech support, it was 100+ hours with tech support, network support, backend support, everything, over many months, close to a year.
In the end, the carriers were frustrated with all the time and effort they had to put into Microsoft devices, and one by one they all dropped them from their device lineups. And VERY few people at the time went out and simply purchased their own phones.
"What we’ve learned about electrification is it’s actually not about propulsion systems; it’s really about what you can do outside of the propulsion system, and also the software," said Farley. Note that the first generation of Ford’s EVs were “analog products” and that Ford intends to ensure that its future products would be digital.
“That’s why at Ford we’ve decided in the second-generation [EV] product to completely insource electric architecture,” Farley continued. “To do that you need to write all the software yourself—but just remember car companies haven’t written software like this, ever. They’ve never written software. So we’re literally writing the software to operate the vehicle for the first time ever.” Ford CEO Explains Why Legacy Automakers Take Forever to Issue OTA Updates
The future of automobiles is the software-define vehicle (SDV). Wards Auto has ranked the major companies by their progress with SDV.
