Which car firm will dominate the future? Tesla and its BEVs or Toyota with its hybrids? Schalk Cloete looks at the cost reductions coming down the line. He says that the hybrids have many more improvements on the way, whereas in terms of performance and efficiency the BEVs are already reaching their peak. Though further and considerable progress in battery technology is coming, it will benefit both. For city driving both will rely on battery power and have similar fuel costs. For distance, Cloete has modelled electricity and oil prices to show the hybrids will cost less to run in most realistic scenarios. And what about the economics of emissions? The carbon intensity of a grid determines how emissions-friendly batteries are. But even with zero-carbon electricity, the CO2 price will have to reach $250/ton before Toyota’s RAV4 fuel costs exceed that of charging Tesla’s Model Y, says Cloete. And many low-carbon fuel options (biofuels, hydrogen, ammonia) will enter the market, benefitting the hybrids, long before such CO2 prices are reached. The evidence is favouring the hybrids, concludes the author.
First, what do the markets think?
Tesla has recently overtaken Toyota to become the world’s most valuable automaker. And the rally did not stop there. Following an incredible 600% gain, shown on a quarterly basis below, Tesla is now worth twice as much as Toyota.
When considering standard financial metrics, this valuation looks downright crazy. Tesla has not shown significant revenue growth for two years, and revenues remain an order of magnitude smaller than that of Toyota (aside from the Covid-19 drop in the most recent quarter).
Tesla has now achieved four consecutive quarters of profitable operations, aided by over a billion USD in regulatory credit sales (the profit/credit ratio is 0.46). Without regulatory credits, automotive gross margins remain steady at about 18%, which is low considering Tesla is a premium brand that skips the middleman (dealerships).
No, Tesla’s astonishing valuation is linked to market expectations of wild future profits. Given that 85% of Tesla’s revenue comes from its automotive business, much of this rally relies on two assumptions:
- Battery electric vehicles (BEV) will see massive and sustained growth.
- Tesla will strongly outperform the BEV competition.
This article presents a critical evaluation of the first point by contrasting the fundamental value propositions of BEVs against that of hybrids and plug-in hybrids (PHEV) championed by Toyota. The second point has little to do with fundamentals and is not within the scope of this article.
The Tesla Effect
Tesla’s meteoric rise is a result of the incredible speed at which they boosted the attractiveness of BEVs. Innovation certainly played a major role, but arguably the biggest factor is the bravado of Elon Musk and his management team. Basically, Tesla established a virtuous cycle in which they developed desirable cars and an appealing vision that attracted many dedicated fans willing to pay big money to be beta-testers for Tesla products.
Getting paying customers to beta-test Tesla BEVs
This large pool of paying test subjects allowed Tesla to quickly roll out products with significant defects and fix the issues based on user feedback. Large pre-order deposits for the Model 3 and ongoing sales of the pure-profit $8,000 “full self-driving” option (that is nowhere near the level 5 autonomy its name suggests) are additional benefits of this strategy.
As a result, Tesla greatly accelerated the development of BEVs in terms of performance, efficiency, battery cost reduction, and charging technology. Meanwhile, traditional automakers were advancing conventional internal combustion engine (ICE) and hybrid technology at the usual incremental rate.
Thanks to this dynamic, I believe that BEVs have already overtaken ICEs and hybrids in the race to their ultimate potential. In terms of performance and efficiency, BEVs have little left to gain. Battery costs have considerable room to fall, but absolute annual cost reductions are slowing and will have trouble compensating for subsidy phase-outs. In many markets, the low-hanging fruit of fast-charging has also already been picked.
On the other hand, ICEs have plenty of potential remaining to piggyback on electric drive advances towards substantial efficiency and performance gains. In fact, hybridisation allows ICEs to derive similarly large benefits from further battery cost reductions and performance improvements to BEVs.
To support this claim, let us take a closer look at the latest offerings from the world’s two largest automakers (in terms of market cap).
Which has a better future? Tesla Model Y vs. Toyota RAV4
Back in 2017, I wrote an article about the technological potential of hybrids. Since then, Toyota has dutifully delivered on two of my expectations from that article:
- Hybrids will become better to drive than conventional ICE options.
- Hybrid systems will be implemented as electric all-wheel drive (AWD).
The RAV4 offers a good example of these benefits. Relative to the AWD RAV4, the electric AWD hybrid offers 8% more power (with even better low-end performance due to instant torque from electric motors) and 33% better fuel economy for only $1,000 (3.7%) higher sticker price.
Recently, the RAV4 Prime PHEV has taken another big step forward. In this model, Toyota has capitalised on the larger battery pack by greatly boosting the power of the electric motors. The result is a 0–60 mph time of 5.7 seconds, which is very close to that of the long-range Tesla Model Y.
In terms of real-world fuel economy, the Model Y achieves about 90.7 MPG (225 miles from the 75 kWh battery pack, accounting for 90% charging efficiency), whereas the RAV4 prime achieves 39.1 MPG in hybrid mode and about 90 MPG in electric mode.
Since the RAV4 Prime and the Tesla Model Y achieve similar performance, we can fairly compare their fuel costs. For conventional city driving, the 40 miles of all-electric range will be sufficient for almost all daily trips, giving the RAV4 Prime identical fuel costs to the Model Y.
When it comes to longer trips, where the Model Y must rely on fast charging (currently 0.28 $/kWh), fuel costs will be more than twice as much as with the RAV4 when oil costs $50/barrel, as shown below.
Thus, performance in the city will be essentially identical, whereas the RAV4 will be much cheaper on road trips (in addition to the convenience and freedom afforded by 567 miles of gasoline range that can be topped up anywhere in only a couple of minutes).
This means that, for fundamental competitiveness, BEVs will have to ensure that a 4x larger battery pack is considerably less expensive than the added ICE in the PHEV. This is a tough ask, given the considerable room left to run before reaching the fundamental potential of hybrid powertrains.
Hybrid Drivetrains: many improvements still to come
Toyota has now started rolling out the hybrid benefits of superior performance and cheap all-wheel drive, but there are many improvements yet to be realised. Here is a list of my top five expectations:
- Larger battery capacity in conventional hybrids, facilitating a large increase in the power of electric motors and a substantial downsizing of the ICE to reduce cost and improve driving dynamics.
- Standardised use of increasingly intelligent systems to optimise hybrid drivetrain energy management to cut fuel consumption by about 20%.
- Greatly simplified ICE transmissions that contain only a few higher gear ratios, relying on the electric motor for all low-speed driving.
- Large gains in ICE efficiency and environmental performance via advanced compression ignition technology. Implementation of such complex engines will be much simpler in hybrid systems that allow the ICE to operate predominantly under steady conditions.
- The gradual introduction of waste-heat recovery systems.
Following these gains, a hybrid will have similar or better (due to lower weight) driving characteristics to a BEV, and similar or lower costs than a conventional gasoline vehicle. In addition, I expect that technologically mature BEVs will only be about 50% more efficient than technologically mature hybrids, making hybrids cheaper to fuel in almost all circumstances.
But there are several additional factors that drive so many smart people to predict the imminent demise of the ICE. Let us now take a critical look at the four most important ones.
The Million-Mile Battery
There has been much excitement about Tesla’s million-mile battery, although this was not mentioned at the recent battery day. However, this development is much more interesting for PHEVs than BEVs.
Very few people will drive their cars anywhere close to a million miles, but, when placed in a PHEV with a 5x smaller battery, such technology will allow for 200,000 miles of all-electric driving, which is more reasonable. In addition, the much smaller battery of the PHEV will also greatly reduce potential challenges with battery materials as electric drive scales up.
If the need ever arises, million-mile hybrids will soon emerge. As hybrid technology improves, the load on the engine will grow smaller, allowing for optimal steady-state operation under almost all driving conditions. The superior reliability and lower depreciation of hybrid cars is already demonstrating the long-term benefits of the synergy offered by hybrid drivetrains.
Driverless Taxis
One application where a million-mile BEV makes sense is autonomous taxis. However, there are several issues with this scenario. First, the technical challenge of safely and efficiently operating a large fleet of fully driverless taxis in a generalised urban environment is huge. After first writing about this in 2016, I remain of the opinion that full autonomy will take considerably longer than proponents suggest, and that benefits will be minimal before almost all vehicles on the road are fully autonomous.
Second, even when we finally reach full autonomy for all vehicles, the benefits of driverless taxis are questionable. A recent study found that driverless taxis will have trouble competing with privately owned cars. This study came to a similar conclusion, also highlighting the importance of cleaning costs in driverless taxis and the fact that customers will have to be monitored by video to ensure they behave appropriately. In urban centers, public transport remains cheaper.
Third, when considering the timescales required for achieving full autonomy in all vehicles, I think the biggest competition will come from virtual mobility and small electric vehicles that will thrive in cities increasingly designed for people instead of cars. Even without accounting for the large quality of life benefits of living in an environment with minimal cars and maximal green spaces, the economic benefits of these options are enormous. Hence, I believe that car traffic will increasingly shift to highways, where the ICE will benefit most.
Grid services for charging
One option for effectively using a privately owned BEV with a million-mile battery is to supply grid services by charging when electricity is cheap and discharging when electricity is expensive. However, executing this strategy in a way that never inconveniences drivers will be very difficult, especially when balancing fluctuating wind and solar power.
Solar power features attractive daily regularity for integration with BEVs, but the problem is that cars will need to be charged in daytime, interfering with use patterns, requiring many costly public chargers, and demanding expensive grid upgrades to handle increased peak system load. Stationary batteries installed at carefully optimised locations to minimise transmission and distribution grid capacity may well be more economical than BEVs for this purpose (and will certainly be much more practical).
BEVs fit very well with a baseload (e.g., nuclear) power system where highly convenient and predictable grid-friendly charging can happen every night. However, this gives minimal opportunities for discharging, meaning that PHEVs will be a better use of million-mile batteries in such a scenario.
CO2 intensity of grids’ impact on emissions
Climate change impacts of BEVs depend heavily on the CO2 intensity of grid electricity. The figure below gives the equivalent efficiency that the RAV4 needs to have to match the emissions of the Model Y. The top of the blue band is the current RAV4 efficiency, and the bottom is the expected long-term efficiency.
Clearly, hybrids will have similar or lower emissions to BEVs in the largest developing car markets for many years into the future (e.g., China’s CO2 emissions intensity is about 840 kg/MWh). Higher biofuel blend ratios and hybrid efficiency gains mean that BEVs will require considerably cleaner electricity to match hybrid CO2 emissions over the coming years.
In regions blessed with cheap and clean electricity, PHEV market share will increase relative to regular hybrids. These PHEVs can drive mostly on electrical power and still use gasoline for longer trips.
…and with a zero emissions grid?
But even if we assume the Model Y from the fuel cost comparison presented earlier is charged with zero-carbon electricity, the CO2 price will have to reach 250 $/ton before the RAV4 fuel costs exceed the cost of charging the Model Y with 0.28 $/kWh electricity from fast chargers. Many low-carbon fuel options (e.g., biofuels, hydrogen, ammonia) will enter the market long before such CO2 prices are reached.
Conclusion
Following this analysis, I remain convinced that Toyota is on a sound strategic path with its balanced hybrid and PHEV strategy. Pure BEVs have an important role to play in the future transportation system, and Toyota will do well to include a few BEV models in their line-up, but the total BEV dominance suggested by Tesla’s valuation seems unlikely.
With the RAV4 Hybrid and RAV4 Prime, Toyota has finally managed to bring the electric drive performance benefits to its hybrid offerings. We can expect this development to proliferate through the rest of Toyota’s line-up, upgrading their brand image from just being reliable and efficient, to being fun as well. I’m looking forward to seeing how quickly they manage to implement the other hybrid drivetrain advances discussed in this article.
***
Schalk Cloete is a Research Scientist at Sintef
Roger Arnold says
I don’t have strong feelings in re to the BEV vs PHEV debate, but FWIW, here are some comments. First, regarding stock valuations, you write:
Yes, but it’s the second point that’s critical to Tesla’s high market cap. Just when it was lookin like the competition might be starting to catch up, Tesla moved the goal posts. Rightly or wrongly, Tesla’s vertical integration and focus on innovation for cost reductions greatly impressed investors and analysts.
I dunno. Price reductions to purchasers may be slowing, but manufacturing cost reductions seem to be moving right along. Tesla can now afford to raise its margins. At least that’s what the market seems to believe.
I think Musk’s view that “the factory is the product” recognizes a deep economic truth: given the will and adequate investment, the marginal cost of production for anything can be brought close to zero. Marginal production cost is labor, and with automation and rapidly advancing AI, the labor content for production can be arbitrarily reduced. That applies even for material inputs. Materials are the produced output of labor. It takes a high degree of smart vertical integration to realize the cost reductions quickly, but that’s what Tesla has going for it.
Not ICEs per se, but for PHEVs, I’d agree. (With some reservations, of course. This is me, always picky.) Your point about million mile batteries being of greater relative value for PHEVs than for BEVs is well taken. The smaller batteries in a PHEV will cycle more rapidly than those in a long range BEV, so cycle lifetime is more important.
I’d also agree that there’s room for gain from smarter battery management in PHEVs. With road maps and navigation systems now standard on all new vehicles, the control system should be able to anticipate upcoming power needs and adjust its battery charge-discharge strategy accordingly. It should try to begin any grade ascent with a high charge level, and draw on its batteries enough for regenerative braking to be used on the downgrade. It’s always a minor annoyance for me on my Prius to see the engine working to bring the hybrid battery up to full charge just before I begin a long downgrade run. Non-regenerative braking is always energy wasted.
I’ve long been puzzled by PHEV manufacturer’s failure to adopt what seems to me the biggest potential efficiency improvement. That would be a small, ultra-efficient constant-speed, constant torque engine feeding a serial hybrid drive train.
I understand why that wasn’t done for the early Prius models; the battery technology wasn’t up for it. It couldn’t store enough energy or deliver enough power. Toyota’s “synergy drive” was a clever work-around for those limitations, but it makes the Prius a relatively conventional ICE vehicle with an electrically boosted CVT. Most of its efficiency gain came from elimination of idling at stops and inefficient low-throttle operation in slow traffic. But once you have the electric motors and batteries sufficient for pure electric-mode operation in a PHEV, why continue with a parallel-hybrid design?
Luke says
Well PHEV are far worse than they seem.
https://www.isi.fraunhofer.de/content/dam/isi/dokumente/cce/2020/PHEV_ICCT_FraunhoferISI_white_paper.pdf
in short: they use 2-4x more fuel as expected from NEDC
there are even stories from company leases where the electric cable was still sealed in the original plastic after the lease
car manufacturers love these cars because they can still produce and sell ICE plus electric on top which makes a more expensive car that does not seem that way because of incentives (direct and indirect because of taxes tied to emissions etc.)
on top they are clean according to the lab test and therefore good for the average CO2 emissions per km in europe so they can avoid emission fines
on another note: yes 600 miles of gasoline range sounds great – unless you want to drive that in one go
i think most people would like a oner or two 30 minute break on such a long trip which is enough to charge a modern ev so they are no actually behind even with a 200-300 miles range
“4 times more battery capacity for less than an ICE” sound hard
but assume 40-80 kWh battery so that means 30-60 kWh extra cost
tesla wants to get to 50$ per kWh, maybe a 100$ sale price ?
that means anywhere from 1500-6000 $
the ICE motor is not that expensive, but you also need: a bigger cooler, a clutch, a transmission, and an ever more expensive exhaust after treatment (i heard in a diesel these system cost around 2000$ alone)
all these system need space and make it more complex which adds cost for the consumer as well
so to me it does not seem that hard to achieve and it simplifies the design a lot
Schalk says
Yes, these are interesting times for Tesla (even by their standards). I think Europe is a good test bed for the competition element. Over here in Norway, Tesla’s share of the BEV market has dropped to 10% averaged over the last year from 25% over the prior year. But the European markets are highly impacted by the huge electric car incentive provided by the emissions standards that set BEV CO2 emissions to zero and count them double in the average fleet CO2 calculation, so it is tough to analyze.
It would be very interesting to see a breakdown of the cost components of the average Tesla car. As a rough estimate of the current labor cost, we can give each of the 10000 Tesla employees at Fremont a $70000/year salary and work on a 500000 car per year production rate, arriving at $1400/year. The average Tesla cost $45000 to produce in 2019, so this is about 3% of the production cost. Not nothing – but no gamechanger if it can be slashed in half. I don’t know how many other workers there are along the full value chain though.
The problem with serial hybrids is all the energy conversions required. For steady state driving, mechanical energy must be converted to electrical and then back to mechanical. This also requires an extra generator and a larger electric motor than the parallel hybrid, cancelling out the savings in the ICE transmission. For variable speed driving, a large fraction of the energy must also cycle through the battery, adding more conversion losses. The battery itself must also be considerably larger to support the more powerful electric motor. To avoid these issues, I believe parallel hybrids will stay, but the energy management systems will become significantly smarter to maximize the time the ICE spends operating in its ideal efficiency window, while directly connected to the driveshaft.
Alex Bruce says
“… in terms of performance and efficiency the BEVs are already reaching their peak.”
This is an utterly astounding statement.
The only justification given for this pontification is that:
“This large pool of paying test subjects allowed Tesla to quickly roll out products with significant defects and fix the issues based on user feedback.”
and …
“Thanks to this dynamic, I believe that BEVs have already overtaken ICEs and hybrids in the race to their ultimate potential.”
The author simply “believes” this statement into existence like Bertrand Russell’s teapot.
Beware Amara’s Law (http://www.rationaloptimist.com/blog/amaras-law/). The development of BEV, I “believe” is only just getting started. Watch this space.
The author himself predicts: As hybrid technology improves, the load on the engine will grow smaller.
What is the logical conclusion of this process of load growing smaller? The conclusion is a steady reduction to zero load – i.e. no engine at all.
Schalk says
I did not think the statement about BEV efficiency and performance would be controversial. EV efficiency should be around 80% now, including battery charge and discharge losses, inverter losses, and electric motor losses. There is simply very little room for further improvement. In addition, BEV acceleration is already more than the average person needs. What additional efficiency and performance gains do you expect?
Thanks for the reference to Amara’s law. It is certainly an interesting observation and I will keep it in mind for the future.
The load on the engine will grow smaller with increased hybrid drivetrain sophistication because it will take less responsibility for hard acceleration events (handled by the electric motor). Ultimately, the ICE is best for steady state driving, which actually needs quite modest horsepower. It cannot reduce below that point.
Alex Bruce says
What additional efficiency and performance gains?
– Increased energy density for batteries (power to weight). Do you know what the limit will be?
– Improved green credentials for batteries. For example C4V has a battery with chemistry that has now been show to have 87% less dirty energy per kilowatt hour versus comparable batteries (https://www.medianet.com.au/releases/190621/). I expect this to increase, while fossil fuels will still be dirty fuels.
– Renewable energy is becoming cheaper. Cheap green electricity is flooding into the market. I expect this trend to accelerate. Filling the electricity “tank” will get rapidly cheaper, while filling the petrol tank will not.
Perhaps we have different perspectives on BEV efficiency. Your 80% is about drive train. My view is broader all the way back to the electricity production.
Schalk says
Ah, OK. It seems like it’s just a difference in our interpretations of efficiency.
As to your other points.
– Lithium ion batteries are approaching their energy density limits. Other battery chemistries can certainly improve on energy density, but they face all sorts of other challenges. Incredible amounts of money is being used for battery research nowadays, so hopefully these challenges can be overcome.
– As is evident from the last graph in the article, the battery pack carbon footprint is not much relative to the electricity carbon footprint. Further gains here will not change the picture much.
– Yes, renewable energy is getting cheaper, but balancing this renewable energy without fossil fuels will be very complex and expensive. Please see this paper of ours: https://www.econstor.eu/handle/10419/222474/. I add the pre-print link, but the paper has just been accepted for publication the International Journal of Hydrogen Energy.
Pavelll says
Thanks for very interesting article, Schalk! I admire Elon for his achievments and style of leading companies … but I have bought Toyota Corolla Hybrid 😉 It has more power and half consumption comparing to my previous Subaru Outback – fantastic! 🙂
Regarding the new trend of more and more powerful Toy hybrids today, one should mention the Toyota CEO Akio Toyoda who likes and participated in car races – one more quaranty that this trend will go on.
Schalk says
Did you consider a BEV in your purchase decision? It would be interesting to hear at what point will you choose a BEV over a hybrid?
Hicham says
Good Luck to Tesla and Toyota
Roland Bavington says
There is a false equivelence in this piece.
The author equates miles done in a hybrid on fossil fuel with miles done in a BEV on Fast Chargers (for longer than average trips). This is not the same thing.
A PHEV uses fossil fuel as soon as it is above a certain speed or when the battery is exhausted, usually once it clears the city limits, for most cities thats just a few miles.
A BEV only needs to use a fast charger if it uses up all of its overnight charge. The Tesla Y has a range of 300+ miles.
So there is a disconnect here, at some point north of 300 miles, prehaps as much as 600 miles a PHEV might be cheaper to run but realistically most people will drive perhaps 500 miles a day max, stoping at a hotel and recharging overnight again at a cheaper rate than on fast chargers, at some it’s free.
Schalk says
Yes, this is true, but the point stands that PHEVs will have the same fuel costs as BEVs when BEVs are charged at home, while BEV fuel costs will be much higher when charged at fast chargers.
I should also add that the practical road-trip range for the Model Y LR could well be below 200 miles. As reviewed in the linked video in the article, the real world range of the Model Y is about 225 miles. Also, when doing fast charging, one would want to leave some safety margin before reaching the charger and will rarely charge to 100% (because it takes too long). Thus, one could bargain on about 80% of the total range between fast chargers (about 180 miles).
Always finding a cheap overnight charger while on a road trip will also be tricky. If this is any hassle, people will prefer charging at the fast charger. Also, when it can be found, these chargers will be more costly than at home. Not as expensive as fast chargers, but still more expensive than gasoline from $50/barrel oil.
These are important dynamics to consider in a future world where telecommuting and small electric vehicles take over more and more transport in cities increasingly built for people instead of cars. In such a world, the car will be increasingly relegated to the highway, where the ICE is in its element and the high energy density of liquid fuels is most valuable.
Roland Bavington says
Its no secret that there will continue to be edge cases where ICE vehicles shine but these will be edge cases and for the vast majority of people, if a car is cheaper quieter and more convienient 360 days of the year no one is going to mind much if it is as expensive or even more expensive on the few days they need to travel longer distances.
Poorly thought-out pieces like this are a gift to climate deniers. You have some interesting things to say but I wish you would withdraw it and give it a re-write.
Schalk says
What I’m saying here is that PHEVs will have the same running costs as BEVs on most days (electric town driving), combined with lower running costs, greater convenience, and more freedom on road trips.
Hence, it’s the same on most days and better on some days. Please be more specific about what is so poorly thought-out.
Roger Arnold says
Regarding the cost of fast charging, I don’t question that the quoted $0.28 / kWh is in fact the current price in some places. I have to say, however, that it sounds outrageously high going forward.
Fast charging stations are best supported by large station batteries fed by low cost surplus power. There is a small loss of efficiency in the two-step trip from grid to charging station battery to customer vehicle (as opposed to direct from grid to customer vehicle), but it’s not much. The Tesla PowerPack is spec’d with a round trip efficiency of 92.5%; that’s after an AC to DC conversion coming in and a DC to AC conversion going out. For a fast charging station, the DC to AC out is bypassed. Charging is DC to DC, from station battery to vehicle battery. So say 93.5% efficiency, grid to customer vehicle. The station battery could be expected to turn over several times per day, so amortizing its capital cost over 10 years would add only a cent or two per kWh to the cost of charging.
Since the charging station would be drawing mostly cheap surplus power from the grid and would allow the station to avoid demand charges for a higher capacity grid connection, I’d be very surprised if it didn’t pencil out as profitable at a cost as low as $0.08 per kWh to customers. Then consider that that fast charging stations are usually co-located (at least in the U.S.) near convenience stores, restaurants, malls, or tourist attractions and are partially subsidized by the businesses that will profit from the increased flow of casual customers. In light of that, it’s quite plausible that fast charging at commercial stations could end up cheaper than home charging.
Schalk says
Well, with an 8% discount rate, the levelized cost of a 5000 cycle battery costing $200/kWh (fully installed) doing two 70% charge cycles per day with 10 cent/kWh electricity is about 9 cents/kWh. Add to that the 10 cent/kWh (generation and grid) electricity that is consumed, and you arrive at 19 cents/kWh. In a baseload power system with predictable low-cost electricity at night, it might be better to just take one night-time charging session, maybe at 6 cents/kWh, which results in a slightly lower total of 16 cents/kWh. However, this will be much more complicated in a VRE-dominated power system.
Then there is the cost of the fast-charging infrastructure (which is used at a very low utilization rate – the faster the lower) and the space (the most convenient locations will be quite expensive). Given all this. I don’t find it hard to understand the 28 cents/kWh price. It seems to be at a similar level for different fast-charging companies around the world.
Roger Arnold says
Yes, sort of. $200/kWh and 5000 cycles is an estimate for the 21700 – based packs that Tesla has been producing for the Model 3. Speculation for the 4680-based packs announced on “battery day” is sub-$100 and probably 10,000 cycles, within two or three years after experience tweaks and production ramp-up. That’s at least the ballpark that Musk is aiming for. It’s the basis for all the talk about the immanent demise of ICE vehicles. Most speculators seem to accept it. They’ve evidently grown tired of losing money betting against Tesla.
I could also argue that your assumed 8% discount rate is too high. True, 8% is commonly used for LC estimations, but for low-risk investments in the economic environment that has prevailed for the last 12 years, it’s questionable. Recent tenders for solar and wind farms seem to show that developers are willing to put money into projects where the cost of capital pencils out to be close to zero.
I’m further assuming that the large batteries at charging stations will be aggregated to provide dispatchable generation and demand regulation to the grid. Their transactions with the grid will all be on the wholesale market. Finally, don’t dismiss the value of a fast charging station to neighboring businesses. Many in the US pay nothing for the land they occupy. In some cases, businesses even pay the operators to install fast charging stations at their locations.
Perttu says
Thanks for very thought provoking and well researched article. I
mostly agree with your findings but I find your conclusions to miss
the bigger picture. I hope you find my comments useful regardless.
How far future are you looking at here? I mean, how long do you think
cars can emit CO2 freely to the atmosphere? My pessimistic (?) view is
that by 2035 it will be very clear to most that we are more or less
losing the game and drastic corrective actions are badly needed. I
assume that by then CO2 is actively captured from atmosphere and
deliberatly putting it there is strongly discouraged. That would be 15
years from now.
So to me it seems that all cars emitting CO2 are on their way
out. They will be replaced by EVs even if they are not “better” in
every way. So how wise it would be to invest in a ICE maker now? Even
with the best of hybrids Toyota would still only be the biggest fish
in a drying pond. And it would be smaller than now in absolute
terms. We’ve already seen Peak Toyota – why would anyone invest in it
anymore?
But we probably haven’t seen Peak Tesla yet 🙂 And they are on to
something when focusing on manufacturing. Getting battery costs down
is not the only way to cheaper EVs. With less mandatory stuff to put
in means you can mold the car differently. You can mold the car to be
easier to manufacture for example. Volkswagen has come to this same
conclusion and decided to build its EVs in separate factories where
only EVs are manufactured. Having ICE, PHEV and EV versions of the
same car on the same assembly line ties your hands – you can not
improve your EVs and their tooling independently (and as fast as you
really should). Also Tesla 3/Y:s super simple dashboards are not
simple only because of styling. They are probably super simple (and
cheap) to manufacture. And again Volkswagen is following Tesla’s lead
(I’ve only seen ID3:s dashboard, though).
I think Tesla’s self driving tech is more important as a new way of
monetizing car manufacturing than technically (at least for now). The
point is getting people to give them money repeatedly instead of just
doing a one-time purchase. Just ask Microsoft how good it feels 🙂 So
Tesla’s (car) plan is to build a simple skateboard battery computer
with wheels and a big screen in the middle as cheap as possible. Then
make money with software. Making (large batches of) physically
identically cars and then differentiating them with software (f.e
unlocking battery capacity *that already exists in the car* remotely)
is similar to what IBM has been doing for decades with mainframes and
Power servers. Here all other car manufacturers (even new EV
manufacturers) are seriously behind.
Porsche sales figures for 2020 show how Taycan (electric) has badly
eaten Panamera (same size ICE/hybrid). That demonstrates the problem
with ICE car manufacturers starting to make EVs – when selling an EV
they are mainly cannibalizing their own ICE sales. Doubling the
development costs for the same amount of cars sold doesn’t sound like
a winning strategy. Again, Volkswagen seems to have clearly decided to
eat *other* ICE manufacturers’ sales before their own ICE sales
dive. And Toyota? Making better and better hybrids and then World
domination?
It seems that “market” believes Tesla has more plausible strategy than
Toyota even though Toyota’s products may be better *now*.
And finally the alternative fuels: as a car buyer you want to buy a
car that can buy energy from as many providers as possible. And as
energy provider you want to sell energy to as many buyers as
possible. Any car manufacturer takes a huge risk when targeting niche
/ new alternative fuel. EV is so much easier – electricity is
ubiquitious and you can get it from countless providers. Building a
super charger network is nothing compared to building, say, a hydrogen
distribution network. And most importantly, electricity abstracts fuel
away from the picture. As an EV manufacturer you really don’t need to
mind yourself with where the energy to move the car comes from –
that’s buyers’ problem and answers vary in time and place. By choosing
electricity you future proof your business. And that is why Tesla’s
market cap is so much larger than Toyota’s. The other one has a
future, the other doesn’t.
Schalk says
Regarding CO2 emissions, technology-neutral policies working on a CO2 price (combined with the removal of all technology-forcing incentives) will quickly direct the market to the right priorities. A price of €100/ton across the economy will create very rapid decarbonization across multiple sectors, but it will only increase the running cost of a hybrid with 100 g/km emissions by 1 cent/km. Note that passenger cars cause only about 8% of global CO2 emissions, and this share will decline as telecommuting gains in popularity and cities are increasingly built for people instead of cars.
I really hope we have already reached peak car. Cars have many other externalities and socialized costs that are way larger than CO2. The trends of car-free city zones and neighborhoods are really great, and the pandemic-inspired rise of telecommuting will certainly aid this healthy trend.
BEVs are great for the luxury/performance segment where instant torque is highly valued and a large battery remains a small fraction of the total car cost. But more electric Porsches will not solve any of our problems. In fact, one of the silliest things about the huge BEV incentives all around the world is how much of that went to help rich people buy more luxury cars.
Regarding alternative fuels, I think gasoline will stick around for a long time, increasingly blended with biofuels and other synfuels. With the CO2 prices required for rapid decarbonization, it will remain competitive, given it’s modest carbon intensity and high convenience and freedom factors.
Perttu says
> Regarding CO2 emissions, technology-neutral policies working on a
> CO2 price (combined with the removal of all technology-forcing
> incentives) will quickly direct the market to the right priorities.
> …
> Note that passenger cars cause only about 8% of global CO2
> emissions, and this share will ..
You are much more optimistic on humanity’s situation than I am 🙂 I
think that putting a price on CO2 was a good idea 15 years ago. Now it
won’t have the intended affects fast enough – I assume that will be
acknowledged by 2030 (price set to infinity on some sectors). To put
it bluntly – you can not decarbonise anything if you keep putting CO2
into the atmosphere. Atmosphere doesn’t care whether CO2 is from
fossil fuels or biofuels (or generated with solar power from
atmospheric CO2). Atmosphere CO2 level at 420 ppm *there is no
sustainable way of putting more CO2 into the atmosphere*. When/If we
somehow get back to 320 – 340 ppm let’s revisit the
subject then. Currently talk about sustainable CO2 is an accounting
fallacy.
The other thing is this talk about global. Let’s just forget global –
there will not be a global “fix this first” list. Different parts of
the world will proceed in different pace. The faster ones attack low
hanging fruits first (cars, electricity) and try to get a technological
edge, the slower and non-action ones just try to get benefits by
blackmailing. Somehow this mess will result in zero CO2 emissions in
20-30 years …
> I really hope we have already reached peak car. Cars have many other
> externalities and socialized costs that are way larger than CO2.
I thought that CO2 was supposed to be an existential threat to
humanity (civilization, not species). I really don’t see any
externalities or social costs working quite at the same level.
> But more electric Porsches will not solve any of our
> problems. In fact, one of the silliest things about the huge BEV
> incentives all around the world is how much of that went to help rich
> people buy more luxury cars.
I tend to disagree on this also. For once rich people are doing
something useful. Governments just create the money spewing machine
but don’t pick winners – rich people do 🙂 And by channeling money to
“winners” they create disruption, give lobbying power to new players,
weaken the incumbents and make the whole playing field more prone to
change. Just imagine where we would be without Tesla, Ørsted and the
likes.
> Regarding alternative fuels, I think gasoline will stick around for a
> long time, increasingly blended with biofuels and other synfuels. With
> the CO2 prices required for rapid decarbonisation, it will remain
> competitive, given it’s modest carbon intensity and high convenience
> and freedom factors.
I really find this kind of reasoning astonishing. We are not
decarbonising anything if we just replace fossil fuel CO2 with
“sustainable” CO2 while CO2 PPM keeps on rising. We are currently at
420 ppm. It has risen 100 ppms in 60 years. How much more CO2 can we
put in there?
Perttu says
“When it comes to longer trips, where the Model Y must rely on fast charging (currently 0.28 $/kWh), fuel costs will be more than twice as much as with the RAV4 when oil costs $50/barrel, ”
I don’t quite get why you don’t compare fast charger prices (consumer price) with gazoline consumer prices but crude oil prices? Where I charge (northern European capital region) my fast charging price is higher (0.31-0.33 €/kWh) which with my Leaf (average around 17 kWh/100km) makes 17 kWh/100 km * 0.33 €/kWh = 5.61 €/100km. Cheapest prices here today seem to be 1.41 €/l for gazoline and 1.21 €/l for diesel. So with those prices I would get 3.98 litres of gazoline or 4.64 of diesel. With diesel Golf I *might* get 100 kilometers with that amount of fuel but I very much doubt I could get 100 kilometers with 3.98 with *any* pure ICE, PHEV or hybrid.
In all EU countries fuel price is mostly taxes which can rise when oil barrel price goes down. So using barrel prices as an independent variable when trying to estimate consumer benefits / behaviour is a bit moot.
Schalk says
If BEVs take over the world, all those gasoline taxes will have to come from somewhere else. Hence, from the point of view of the economy as a whole, backing out all taxes is the best way of doing this comparison.
Note that I added refinement and distribution costs to the oil costs in the calculation to make sure the comparison is fair.
Perttu says
> If BEVs take over the world, all those gasoline taxes will have to
> come from somewhere else. Hence, from the point of view of the economy
> as a whole, backing out all taxes is the best way of doing this
> comparison.
Ok, that is a valid point. But electricity consumer price at charger
contains at least VAT.
> Note that I added refinement and distribution costs to the oil costs
> in the calculation to make sure the comparison is fair.
Yes, it’s very tricky to model as electricity can be generated with so
different fuel/machinery cost mixes (eg. sun, wind, nuclear, coal,
gas). At least a few Nordic charging networks claim their electricity
comes from sun or wind only. No fuel costs at all. How anything could
beat that?
Alex Bruce says
Hi Schalk,
I thinks it’s fantastic that you have taken the time to respond thoughtfully to all the comments. I’ve followed through and there is no doubt that my financial knowledge is well short of yours.
However, I am still left with the feeling that your article is the equivalent of discussing the financial merits of harnessing a small pony to the front of a Model T Ford.