The Fractured Tipping Point Moat

[SteveG3]

Glad to see everyone jumping into this discussion. As it happens, some family came in from out of state yesterday, so, for the next few days, it's a little tricky for me to post very much, but I'll try to catch up later with replies to the specific points being raised.

For now, here are a few thoughts,

Beyond a way of thinking about Tesla as an investment, I think this thread is leading us to discussing a very big and compelling topic about how a large very probable global event is going to play out,

If it's going to be EVs, how and when does the battery supply happen?

I'd kind of half want to rename the thread that.

I think jhm's statement,

"So the moat is not that others will struggle to build out battery supply. The moat is that Tesla will have the leading edge solution to build out the supply across the industry. Operating a Gigafactory is a commodity business, but designing and engineering Gigafactories is not."

is quite an interesting one to think about. When Elon repeats a phrase often, it's often a harbinger of an as yet unannounced 3D chess piece. Elon has talked about for over a year "the machine that builds the machine." More recently, Elon responded to the question, ~"what's Tesla's strategic advantage?" with, ~ for now automation, over time, being the world' best manufacturer. So what jhm described may be getting at where Elon currently is trying to skate to and find a puck waiting for him... entities like nations with massive oil sovereign wealth, a scale of money Tesla (or the auto industry for that matter) doesn't have as potential partners for the world's best manufacturer with by far the most experience in building massive GFs. Of course, Elon's talk with a leader of UAE this week was very tantalizing, and led to some speculation along the lines of what jhm suggested by people here on TMC (fwiw, myself included). If it does I see a considerably slower timeframe than what jhm suggested, but I'll get to those details later.

I'll also toss in that I think the idea being raised as to how consumers will react if there's a gap between EV supply and demand is quite interesting. I agree about the likelihood of people holding onto ICE longer until there are available EVs, rather than buy a new ICE. I've been thinking there might be a lot more "horse trading" going on, as up and down the all the price points for vehicles, people might sell their used ICE vehicle to upgrade for someone else's used ICE vehicle.

 

[FlatSix911]

What about 3rd party battery suppliers increasing their output

LG is adding production capacity at its battery factory for Chevy Volt, Bolt EV and Chrysler Pacifica

Korea-based electronic giant LG is making most of the components that make the Chevy Bolt EV an electric vehicle, including the battery pack, electric motor, and power electronics, and the battery packs for the Volt. Therefore, its production capacity will determine GM’s total output for what is currently its only all-electric vehicle in production. In what could be a good sign for the Bolt EV and the Volt, LG announced this week an expansion of its production facility in Holland, Michigan.

Last year, the plant had two of the three assembly lines operating 24 hours a day and a fourth line on the way. It had an annual production capacity of 650 MWh, which would allow for only a few thousand Bolt EV or a few ten of thousands Volt battery packs. But the company said that it could rise to 3 GWh at full capacity and the fourth line alone could have already pushed the capacity passed 1 GWh.

 

[RobStark]

BYD sales are in a number of non-Chinese markets. The E6 is the primary model so far for export, but spec's vary by market

The overwhelming sales of passenger cars for BYD is in China, they market in a number of developing economies but don't sell very large numbers there.

 

[jbcarioca]

The overwhelming sales of passenger cars for BYD is in China, they market in a number of developing economies but don't sell very large numbers there.

Yes, not yet.

 

[RobStark]

I think a startup BEV maker might be willing to take the risk of going from breakthrough to installing in cars within five years.

Legacy automakers will want to do 10 years of testing. Liability, loss of reputation etc is too great to take the chance.

BMW readies radical battery technology for 2026 launch | Autocar

BMW expects a breakthrough in battery technology in 2026, by which time it plans to have solid-state batteries ready for production in its models.

The batteries are in development but 10 years away from production, with long-term durability testing cited as a key reason for the delay.

 

[jhm]

PHEVs make sense when batteries are expensive. It allows manufacturers to keep the cost of the car relatively low (smaller battery) while still giving adequate range to the vehicle. Once batteries get cheap-enough, the concept of PHEVs becomes less and less relevant.

Also, it is widely mentioned that PHEV owners graduate to BEVs once they understand.

So, I think PHEVs will be a dying breed as batteries continue to improve and get cheaper. I would venture to presume that the Bolt will outsell the Volt.....

I think PHEVs will fall out of the market when the time comes, as well HEVS. We can watch this play out in the bus market.

Electric Bus Sales Exploding In China (2010–2015)

Notice the chart with sales by BEV, PHEV and HEV. HEVs dropped. PHEVs were still growing into 2015, but BEVs skyrocketed that year, and continued to 116,000 in 2016. I wish I knew how PHEVs fared in 2016. I suspect the growth will level off. A big issue with buses is maintenance cost, and my suspicion is that the diesel motors in a PHEV may simple not be worth the added maintenance.

 

[EinSV]

I think PHEVs will fall out of the market when the time comes, as well HEVS. We can watch this play out in the bus market.

Electric Bus Sales Exploding In China (2010–2015)

Notice the chart with sales by BEV, PHEV and HEV. HEVs dropped. PHEVs were still growing into 2015, but BEVs skyrocketed that year, and continued to 116,000 in 2016. I wish I knew how PHEVs fared in 2016. I suspect the growth will level off. A big issue with buses is maintenance cost, and my suspicion is that the diesel motors in a PHEV may simple not be worth the added maintenance.

According to ev-sales.blogspot, BEV bus sales in China were up from 94,260 in 2015 to 115,700 in 2016. In the same period, PHEVs dropped from 23051 to 19376. So as you suspected, PHEV bus sales now appear to be dropping in China as BEV sales dominate.

EV Sales: China PHEV Buses 2016

 

[jhm]

According to ev-sales.blogspot, BEV bus sales in China were up from 94,260 in 2015 to 115,700 in 2016. In the same period, PHEVs dropped from 23051 to 19376. So as you suspected, PHEV bus sales now appear to be dropping in China as BEV sales dominate.

EV Sales: China PHEV Buses 2016

Excellent. Thanks for finding this. Nice to see confirmation so quickly.

BTW I read somewhere that transit buses have a 12 life. Fleet managers are buying replacement vehicles every year. So if electric buses are performing well in a fleet, you can expect repeat sales each year. My hunch is that Chinese bus makers decided to push on global sales recently. Perhaps that is why the rise in sales in China was not so dramatic this last year. But as a strategy, it makes sense to build up sales globally. It's a long sales cycle especially where adding infrastructure is involved, but can deliver solid sales year after year.

 

[renim]

PHEVs make sense when batteries are expensive. It allows manufacturers to keep the cost of the car relatively low (smaller battery) while still giving adequate range to the vehicle. Once batteries get cheap-enough, the concept of PHEVs becomes less and less relevant.

Also, it is widely mentioned that PHEV owners graduate to BEVs once they understand.

So, I think PHEVs will be a dying breed as batteries continue to improve and get cheaper. I would venture to presume that the Bolt will outsell the Volt.....

that is logical, but

vehicle #1 (Ford F series) is best done by a PHEV
vehicle #2 (Chevy Silverado) is best done by a PHEV
vehicle # 3 (Ram P/U) is best done by a PHEV
vehicle #4 Honda CRV is probably best done by a PHEV
vehicle #5 Nissan Rogue is probably best done by a PHEV
vehicle #6 Honda Civic is best done by an EV
vehicle #7 Toyota RAV4 is best done by an PHEV
vehicle #8 Toyota Corolla is best done by an EV
vehicle # 9 Ford Escape is best as an PHEV
vehicle # 10 Toyota Camry is best as an EV

take a hypothetical $80/kWh cell (coming in next 5 years)
for a USA pickup truck, a 60kWh PHEV is far more pragmatic than a 120kWh EV
(just try towing anything wide and tall, behind a Tesla X or a Outlander PHEV) the efficiency plummets.

anyway, I'll log off now for the next 3 or so weeks

 

[MitchJi]

take a hypothetical $80/kWh cell (coming in next 5 years).

Tesla's pack costs will be under that by the end of 2018!

 

[jhm]

that is logical, but

vehicle #1 (Ford F series) is best done by a PHEV
vehicle #2 (Chevy Silverado) is best done by a PHEV
vehicle # 3 (Ram P/U) is best done by a PHEV
vehicle #4 Honda CRV is probably best done by a PHEV
vehicle #5 Nissan Rogue is probably best done by a PHEV
vehicle #6 Honda Civic is best done by an EV
vehicle #7 Toyota RAV4 is best done by an PHEV
vehicle #8 Toyota Corolla is best done by an EV
vehicle # 9 Ford Escape is best as an PHEV
vehicle # 10 Toyota Camry is best as an EV

take a hypothetical $80/kWh cell (coming in next 5 years)
for a USA pickup truck, a 60kWh PHEV is far more pragmatic than a 120kWh EV
(just try towing anything wide and tall, behind a Tesla X or a Outlander PHEV) the efficiency plummets.

anyway, I'll log off now for the next 3 or so weeks

Certainly most in the auto industry would agree with you. There is a certain machismo associated with trucks and burning lots of fuel. A lot of men in America hang their identity on that.

But this is precisely why it will be necessary for Tesla to get into trucks. They need to break this perception down just as they have done with high performance sedans. I believe it is strategic for Tesla to get into semis first. That is where all this practical stuff about being able to haul load gets worked out. Moreover, I believe the economics may actually be more favorable for heavy vehicles than a light pickup truck. So when Tesla Semis are showing are able maintain the same speed on an up grade as on a down grade all while saving a ton on fuel, there will be little doubt but that electrics make the best tow vehicles. So this breaks down the whole mythology around burning fuels for power. Then Tesla can introduce a pickup that redefines what men want in a truck.

 

[SteveG3]

might take me a little time, but I'm going to try to catch up on the thread some,

@renim,

from your posts #4 and #8 here's where we agree...

-"long term EVs will go to 100% market share"

-the Mitsubishi Plug-in Hybrid Outlander is a nice product (when I researched it a couple of years back, they were selling it for the same price as the regular Outlander. I don't know if that's still the case, but if so, a pretty neat trick)

- we may see large growth of PHEV sales near term (though we disagree with the reason, as I think this will be do to incumbents making PHEVs available in far greater numbers than EVs despite consumers by and large soon preferring long range EVs to PHEVs)

-this PHEV Outlander and the Model S/X are not cross-shopped meaningfully

and, for friendly dialogue, here's where we have very different views

-I don't agree with your assertion that there is more room for improvement in PHEVs than EVs

-While some portion of the market will want range of 1000 km (~620 miles), I think that portion globally is something far far far far less than 50%, and my opening post was about the tipping point when demand for long range EVs will reach 50%. Please bear in mind, in point 1 of my first post, I choose 2020-2022 as the likely timeframe for this tipping point event as I think the amount of time needed for SuperCharging to 80% will be down to about 10 minutes, and longer ranges will be available for each price point. These were 2 of the 3 critical factors I saw needing improvement... so we agree improvement is needed, but I'm not sure whether you agree with me that this will be good enough for at least 50% of consumers.

- As to comparing PHEV Outlander sales to Model S sales, I'm with RobStark and JRP3 in disagreeing with your claim of this being a favorable comparison for the PHEV. As a matter of fact, I'd say the numbers very heavily give the edge to the Tesla. some details in addition to what Rob and JRP brought up:

1. among all brands there are roughly 10 vehicles sold globally at the Outlanders price point as compared to the far less accessible Model S price point (fwiw, I refer to the Model S, as only 3,700 X were sold in Europe through the end of 2016 which is where your data ends). So, if the data you cited is correct, adjusting for this factor, the Model S relatively outsold the Outlander 5 to 1 for the carved out section of the globe you requested. Backing out that dicing up the globe you choose, and just looking at world total sales, the Model S outsold the Outlander by about 1.5 to 1, which means roughly 15 to 1 when you account for the Model S far smaller very high budget part of the vehicle market.

2. this is not only apples to oranges in terms of SUV vs. sedan, and very different price points... the numbers in your comparison are heavily skewed by country specific incentives. you mentioned the Outlander outselling the Model S 10:1 in the UK... were you aware that the UK, with 16% of Europe's population, made up a third of the Outlanders sales in Europe, and the Netherlands, with just 4% of Europe's population also made up a third of the Outlanders sales? This means that in the other 80% of Europe, the far more expensive Model S actually outsold the Outlander (before even adjusting for price point).

3. while I do agree with you that a PHEV drivetrain works better for an SUV, I don't think a long range EV is any less suited for an SUV than a sedan. in other words, I think with an SUV, a PHEV narrows the gap vs. a long range EV in terms of percent of consumers preferring each, but I do not think it will overtake it.

I'm sure there's a good chance I've not changed your opinion on those points we don't agree on. In a few years, we'll see some apples to apples comparisons.

 

[RobStark]

-the Mitsubishi Plug-in Hybrid Outlander is a nice product (when I researched it a couple of years back, they were selling it for the same price as the regular Outlander. I don't know if that's still the case, but if so, a pretty neat trick)

Back in the day Mitsubishi UK had a cheap petrol Outlander but the Diesel and PHEV had the same premium over the petrol.

Now the base is a diesel for 25k pounds while the PHEV starts at 32k pounds. Serendipitously the top spec Outlander PHEVs have the "Model S" trim. Upwards of 44k pounds.

Outlander Models | Hybrid Electric SUVs | Mitsubishi Motors UK

 

[SteveG3]

@SteveG3,

This is an incredibly thoughtful and insightful post. I generally agree with your overall thesis, although I think the timeline in which this all plays out may be accelerated by quite a bit. Here are a few reasons why:

• I believe Tesla is on a path to produce a much higher volume of vehicles in 2025 than 3M -- probably more like 6-10M, if not more if serious BEV competition does not enter the arena
  • Assuming 1M vehicles in 2020 and a 50% annual growth rate thereafter results in 7.6M vehicles in 2025
    • Assuming no meaningful competition for the reasons you discuss, 50% growth rate is likely too low (it will be much higher between now and 2020 in all likelihood and I see no reason why it would slow down as the technological advantages of BEVs over ICE continue to grow)
    • Since these vehicles are likely to be sold in the most profitable markets (high-end sedans/SUVs and pickups), this will put a tremendous amount of pressure on the incumbents who play in those markets
  • BEVs are at the beginning of their life cycle and undergoing very rapid innovation. So the advantages of BEVs over ICE that you outline are likely to become a gaping chasm by 2020 and certainly by 2025
  • I think Gigafactories will produce more battery packs and drive trains (and possibly cars) at much lower cost than people expect
    • In estimating the number of GFs and capital required to transition from ICE to BEV there is a tendency to use the current cost and capacity, but:
    • Tesla's first Gigafactory is slated to have triple the capacity that was originally forecast just a couple years ago -- a massive increase in production efficiency and capacity in a very short period of time.
    • It seems highly improbable that after such massive improvement, progress in GF production efficiency will be frozen in its tracks. To the contrary, given that only one GF has ever even been partially built, I would expect continued rapid innovation and improved production efficiency, resulting in higher capacities and lower costs.
    • Elon realized that the key to Tesla's future is to become the best manufacturing company on the planet, and Tesla will be investing enormous resources in improving the "machine that builds the machine."
    • Elon has said that manufacturing expertise will provide Tesla's main competitive advantage for the next 10 years. This is another element of the moat.
• In addition, BEV's significant cost and other advantages for autonomous vehicles may force the hand of traditional manufacturers sooner rather than later. Autonomy is coming, and it will be very hard to compete selling ICE vehicles.
• If the legacy automakers continue to stall, Elon is not one to sit around patiently and allow a slow transition.
  • As you note, Tesla already has technological superiority by most measures. Once costs drop and fast charging is in place (2-3 years) BEVs will be technologically superior across the board, and the only barrier will be capital.
  • I don't think Elon will allow capital to be a barrier for very long. If needed, Tesla will not have any trouble attracting large amounts of capital, and it should also be able to internally generate tons of cash flow that can be invested in future GFs.

Bottom line: I think your analysis is excellent. For the reasons outlined above, however, the timeline for the transition may turn out to be faster than you propose, with the same basic outcome.

Thanks again for putting this together. Very interesting to think about.

Appreciate that EinSv

Yes, things may go faster than I think... but, I find it very improbable they'll go fast enough to reach 50% EV supply before 2030.

On some of your specific points,

3M or 6-10M vehicles? The word I've heard Elon use for 2025 is a few million vehicles. That said, the goal of 500,000 vehicles got moved up by 2 years from 2020 to 2018 in the Spring of 2016 (though it's not clear if he meant 500,000 built in 2018, or 500,000 run rate by the end of year). At some point banks may be happy to lend Tesla money, but I'm not sure when that is. I may be too conservative in that I basically map that ramp up out on the assumption that they'd be funding growth mostly from retained earnings. Putting money aside, there is still a question of logistics, i.e., how fast can you build multiple factories bigger than any building in the world with billions at stake for each project? I think its worth noting, that Elon himself on one of the calls when asked about GF2, GF3, etc, said roughly, -I think we want to see the first GF built properly before we start building more-. Not overreaching the speed at which they are capable of managing the task of building these tremendously large "machines that make the machines" is even more of an issue, if, as you suggested, they are going to continue to push forward very strongly on re-engineering the whole design of the plant. And, most likely, it's not just Tesla's comfort level with what is realistic, but Panasonic's comfort level as well. Bottom line, if they beat the 3 million I'm estimating, great, but I think it's unlikely they break 5 million. Using my analogy in the first post, I doubt this will impact how the other automakers respond to staying on the cliff or jumping given the ~100 million vehicle size of the market.

More efficient GFs? you make a fair point here... perhaps by GF3 or so, output will be enough for 1.5 million vehicles vs. 1 million for the likely revised GF1 plan. I still see roughly 5 million vehicles as the upper limit. This growing GF building expertise gap may eventually draw incumbents to partner with Tesla, when they do commit to EVs. This is something worth thinking about the overall pace... but I doubt any one them joins up with Tesla before 2025, and I still doubt we get to 50% supply before 2030.

Re Autonomy... yes, this could accelerate EVs some, forgot about this one. I've actually written on the thread below that I think that the EV powertrain may be Tesla's key competitive advantage in winning market share in autonomous ride share. I think cities are likely to select a company or two as sort of a utility for this service, and there will be cities thinking about their air quality choosing EV fleets over ICE (and other cities choosing crony manufacturers of ICE as well). Even if there's no such utility, large amounts of the populations of large cities are concerned re emissions and air quality, as well as global warming. There is a very high probability of tremendous money in autonomous ride share... so, yes, a strong inducement to use the drivetrain likely to be widely preferred in the selection of an autonomous cab ride. That said, there are actually surprisingly low numbers of cabs to service a city. NYC I believe has 15,000 cabs, though an autonomous fleet would be larger, to a great extent replacing car ownership in dense urban areas. Still, don't be surprised if the incumbents design special autonomous EV cabs that they "explain to us" isn't really what consumers buying their own vehicles want, they'll "let us know" the consumers want plug-in hybrids, lols. Good point, and one to watch.

Speculative Tesla owned Uber-like service impact on Tesla's Mission and Valuation?

 

[SteveG3]

Back in the day Mitsubishi UK had a cheap petrol Outlander but the Diesel and PHEV had the same premium over the petrol.

Now the base is a diesel for 25k pounds while the PHEV starts at 32k pounds. Serendipitously the top spec Outlander PHEVs have the "Model S" trim. Upwards of 44k pounds.

Outlander Models | Hybrid Electric SUVs | Mitsubishi Motors UK

helpful & funny (no button for that yet)

 

[neroden]

I expect the historical average growth rates of electric car sales of 50% every year should continue until electric cars are over half the market. I see no information which would cause me to doubt this projection.

 

[renim]

might take me a little time, but I'm going to try to catch up on the thread some,

@renim,
...
-I don't agree with your assertion that there is more room for improvement in PHEVs than EVs
........

- As to comparing PHEV Outlander sales to Model S sales, I'm with RobStark and JRP3 in disagreeing with your claim of this being a favorable comparison for the PHEV. As a matter of fact, I'd say the numbers very heavily give the edge to the Tesla. some details in addition to what Rob and JRP brought up:
.......

3 reason's why PHEVs have more room for improvement over EVs vs 1 reason for EVs have more room for improvement

both PHEVs and EVs benefit equally from improvements in battery energy capacity and cost, but only PHEVs benefit far more from improvement in battery power and (arguably longevity). Improvements to battery power (but not energy) strongly benefit PHEVs.
Since 2012, for the same # of cells, Tesla has upped capacity from 85kWh to 90kWh, and increased number of cells from 90kWh to 100kWh. For the number and depth of significant car upgrades that have occurred (AWD, autopilot 1&2, falcon doors, weight reductions, electronic fusing) battery/cell progress has been maturing, (and we see this at a commodity cell level)

PHEVs benefit from upgrading the power of a cell, the cost per kW has and continues to drop more rapidly than the cost per kWH. That preferentially benefits PHEVs, along with tricks that allow wider SOC ranges for PHEVs (increasing the SOC window from 60% to 90% is a one time 50% increase in range for nil additional gross kWH capacity. compared to a BEV which may increase the SOC from 92% to 94%.

PHEVs also benefit from ICE downsizing, when I try to back calculate ICE costs from Chinese cars, I get a basis similar to
$100/unit +
$100/cyclinder +
$100/ litre+
$10/kW
1.5 factor from supplier gate to retail.

so about $1,500 for a 3cyl 100kW range extender motor (with exhaust and fuel tank, but not transmission or generator)
so I look to developing world focused companies like Geely (Volvo) and Mitsubishi (SAME )
2018 Volvo XC40 – first sighting of future Audi Q3 rival | Autocar
2018 Mitsubishi Eclipse Cross Spotted During Winter Testing

both BEVs and PHEVs benefit from home energy charging and home solar, as well as destination AC charging. BEVs benefit from enroute DC charging, which is mostly just show for PHEVs.

re: Outlander PHEV comparison to Tesla S, is about the tech, not the company. Tesla was/is a far better managed company. Mitsubishi's decision to bring the kei class IMiev to USA but not their SUV PHEV is an excellent example of that.... However, when we look at locations where both are for sale, we see far greater market acceptance of the PHEV, and also new PHEV entrants taking market share growth from Mitsubishi. Compared to Tesla, who effectively only has conventional vehicles to compete against, their new energy competitors are absent.

Even the USA BMW i3 with its stupidly placed, stupid little gasoline tank, and its stupid software settings had a 3:1 ratio choose the PHEV version over the EV version (despite CARBs 50% limit) 19,130 BMW i3 REx Recalled In US Due To Fire Risk...Thanks To Its Gas Extender

and really, for compact class car that doesn't tow, a full 60kWh BEV (ie Bolt) is better than a PHEV (i3 REX).

but for towing, its PHEV all the way, they combine EV traction and gasoline energy density. for the 1 week use a year, thats enough to make the matter simple and settled.

 

[DaveT]

@SteveG3 Hope you don't mind if I chime in with a few devil's advocate-type of thoughts.

First, I don't think making big battery factories is that difficult. It mostly just requires capital. So let's go with this thought more.

Panasonic/Tesla might have a slight edge in auto batteries at the moment but I don't think the competition is that far behind (ie., LG, Samsung) in terms of cost or quantity. If you look at both Faraday Future and Lucid Motors, both probably could have chosen Panasonic as a supplier but ended up choosing LG (for FF) and Samsung SDI (for Lucid). In other words, LG and Samsung likely offered them better deals or at least similar. Now, there's probably a lot of factors involved (ie., quantity discount, etc), but I'm using this example to show that as of yet Tesla doesn't have any super revolutionary battery tech that others aren't able to copy (ie., like breakthrough in super capacitors). Well, at least nothing that Tesla has been public with (it's possible they have some secret endeavors going on).

So that said, if the technology behind the batteries is more or less commonly held, then most of the challenge in building battery supply is being able to afford and pay the money to build very large battery factories.

Theoretically, most of the large auto companies can afford billions to invest into gigafactories. The main reason they're not doing so right now is because they don't see a demand for EVs that justify that kind of investment.

Let's take Ford as an example. They made $10B in profit (pre-tax) last year... surely they could afford to invest $2B a year in gigafactories for the next 10 years. If they did so, and $2B (along with partners) allowed them to have batteries for 500k long-range EV cars, then in 10 years they would have capacity for 5M EVs. If they upped their investment, they could theoretically have batteries for more cars.

To me, the big issue is not whether or not legacy auto makers CAN make big battery factories. In my mind, that's not an issue. They've got a lot of money and they can easily invest heavily in big battery factories. The main issue is rather a demand issue. When will these legacy auto makers see the demand that justifies them to make big investments in battery factories.

I think we can answer this demand issue in a few ways. One, legacy auto makers can wait to see the demand for EVs among their own customers. However, this will take far too long. Another way is legacy auto makers can see market trends and buying patterns and from that conclude demand is there to justify large battery factory investments.

In other words, when/if legacy auto makers see Model 3 reservations holders actually follow through and buy hundreds of thousands of Model 3s, I think this will be enough justification for legacy auto makers to start to plan on making large battery factory investments. Now, at first they don't need to commit to building 10 gigafactories each. Tesla never did that, and these auto makers don't need to do that either. They'll commit to building one and then try to push out a compelling EV to compete with Model 3. If it succeeds, then they will build more.

Therein lies the real challenge. Legacy auto makers need to compete with the Model 3, and they need to do so successfully in order to earn the demand for their own EVs that will in turn justify large investments in multiple gigafactories of their own.

The challenge becomes even more complex as Tesla is not standing still. Model Y will be coming out, a pickup truck. And then maybe even a Gen4 vehicle or vehicles. Every legacy auto maker is going to have to compete with each of Tesla's vehicles and compete well in order for them to justify more than one gigafactory.

Let's look at this in more detail through an example.

So, VW is interested in investing in their own gigafactory. Let's say they start building it and it initially provides batteries for their line of EVs. Now these EVs have to sell decently well enough for VW to need more batteries and to justify more investment in growing the gigafactory. If they can make a compelling EV that competes well against Tesla, then I think they'll sell a lot of EVs and they'll invest a lot of money in building battery factories.

I think when looking at the entire auto industry there could be some auto makers that can make compelling EVs to compete well with Tesla and others that can't. Those that can't will be headed toward a death spiral. But those who will be able to compete with Tesla on EVs, will be able to survive.

So now the question is how many legacy auto makers will be able to make compelling EVs that compete with Tesla? This question is much more difficult to answer. Some might say there's none. Some might say a good handful or more will be able to.

I lean toward thinking that I think we will see a good 2-4 legacy auto makers compete with Tesla well on EVs, and that will result in them building very large battery factories. Or it's possible that Tesla even helps them (ie., Elon feels sorry for Toyota and Mercedes and feels indebted to them, so he offers to supply them powertrains at great prices so they can survive).

Overall, I think there will be a lot of carnage in the auto industry over the next 10 years or so. But I don't think I agree with the main thesis of @SteveG3's "fractured tipping point moat theory", which (correct me if I'm wrong) asserts that building battery supply will be legacy auto makers' biggest problem in competing with Tesla/EVs.

The more difficult task will be competing against Tesla's products.

 

[neroden]

I also lean towards thinking that we'll see 2-4 competitors. At this point I see the most likely list as BYD, BAIC, and LG Chem.

The automakers who are sourcing both batteries and motors from LG Chem are turning into coachbuilders and will eventually be absorbed.

 

[DaveT]

Warren Buffett wrote in his annual report published today:

"I’m not aware of any enticing project that in recent years has died for lack of capital. (Call us if you have a candidate.)"

The way I look at it is the world is flush with trillions of dollars of money in search of good investment opportunities. The scarcity isn't the capital, it's the good investment opportunities.