The last car you’ll ever own
Zed: Edwards. Let's put it on.
Edwards: Put what on?
Zed: The last suit you'll ever wear.
Will your next EV be the last car you ever buy?
During a recent episode of the five hundred mile commute, we had time to cover a wide variety of topics. While B is a lively conversationalist, ChicoDog tends to harrumph a lot in protest at having to bounce along in the back seat for five or more hours at a time. The topic of EVs came up, as it has more often lately.
Our Camry has 205,000 miles on the clock, and while it's still running well, we are aware that the percentages are not in our favour. Lately, we’ve had to do a lot of combustion-engine-related maintenance work, addressing things like oil leaks, sloppy engine mounts and clogged air sensors. This is fair, for a fifteen-year-old car headed for a quarter of a million miles. Casual discussions with auto mechanics suggest that 300k miles is not an impossible target for this model year, with careful maintenance.
Knowing that the need for an additional, newer car is looming, we discussed buying an EV. We've rented several EVs now (Tesla M3, Bolt EUV, Polestar, Kia EV6) and have a pretty good sense of what it's like to own one. With hours to kill, we explored some of the second order consequences of widespread EV adoption. Some interesting ideas showed up.
Let’s start with obvious trends
EVs will soon comprise the majority of car types in the fleets of most countries [1]
Range will continue to increase and charging times will continue to decrease. These are two sides of the same coin. The faster EVs can charge and the further they can go, the more difficult it will be for traditional car manufacturers to offer any reason why people should stick with an ICE vehicle [2].
The EV charging network will continue to densify. EV charging stations will continue to be built, and charging concerns will fade. Focus will shift from “what if there isn’t a charger when I need one?” to “which of the n available chargers provides the best-priced electrons?”
People will continue to install solar panels, increasingly paired with battery storage. Just today, a Chinese auction closed with the cheapest utility-scale batteries ever [3]. Even the EIA (lol [8]) thinks we’ll continue to install panels.
VPPs using parked EVs and home batteries will become ubiquitous. This is already happening in California.
These are hardly controversial projections. You can simply follow the trend line on charts from the past few years, even assuming linear growth, which is very conservative. Much has been written about these probable outcomes and I can direct you to sources that document these trends.
Less obvious trends, but still well within the realm of foreseeability
Gas stations will begin closing, or converting to charging facilities [4]
Used ICE cars won’t be worth shit. The sale/resale/maintenance ecosystem that surrounds ICE vehicles is going to change dramatically. EVs last longer, much longer. There is strong evidence emerging that BEVs specifically (Hybrids less so) are lasting at least as long as ICE vehicles. Nobody is going to want your old ICE car especially when gas stations become scarce.
Governments will begin implement cash-for-clunker style programs to get ICE vehicles off the roads.
Grid-smart appliances. Utilities will begin offering rebates to consumers installing grid-responsive appliances.
Excess solar build-out. A trend recently observed in Australia and likely to become commonplace everywhere is that new domestic rooftop solar installations are scaled well beyond the needs of the home, essentially turning every house into a mini power plant, producing more energy than it consumes. It is also reasonable to assume that a similar effect will hold for batteries — once every home has sufficient battery storage to cover a day or two (or three) worth of energy, the surplus will go back into the grid at very low (or even negative) prices.
A century-long head start
Internal combustion cars have been available for purchase since 1891 or 1886 if you include Benz’ first, limited offering and 1896 if you focus on the US when Duryea Motor Wagon Co started making gasoline automobiles.
Whomever you choose as the commercial progenitor, it doesn’t change the fact that ICE has had a more than a hundred years to develop. BEVs on the other hand, have had thirty. Interestingly BEVs and ICE share the problem of early development being handicapped. In the UK, the Locomotives Act 1865 required many self-propelled vehicles on public roads to be preceded by a man on foot waving a red flag and blowing a horn, effectively kneecapping development. Similarly, EVs have long suffered from underinvestment and paper compliance, with ICE manufacturers uninterested in anything but the comfort and profit of the status quo.
I’m simplifying here of course. EVs haven’t had to develop chassis layouts or disc brakes or traction control or seat belts ab initio — they simply use those existing technologies. Treating EVs as a de novo new approach to transportation is disingenious.
While startup EV companies have been happily nibbling away at the lunch of the traditional ICE makers, they have also had to contend with the predictable onslaught of opposition propaganda from those incumbents. Backed by the Oil and Gas industries, there’s plenty of money for disinformation campaigns.
An early favourite was the “EVs pollute more than ICE due to the dirty nature of the electricity they use” line. This one was quickly and easily debunked by pointing to the steady (inexorable!) greening of grids everywhere. With each incremental coal plant retirement, that objection became less true, until, for many countries, it became laughable. Soon to follow into the debunker was the idea that the mining of lithium and other minerals is as dirty as oil extraction, or worse, immoral due to child labor and human rights miscarriages in the extraction of cobalt, for example (new EV batteries don’t use cobalt etc).
For each of these objections, a simple search for relevant research quickly yields concrete facts that show incontrovertibly that ICE is, and will unavoidably always be, more polluting than electric drivetrains.
Things are getting interesting
Now that the world has mostly accepted that EVs are just better, and we’re settling in to the adoption phase, things are starting to get interesting. As BEVs begin to rack up serious miles, we are able to start looking at lifetime data to understand what kind of longevity we might expect from a battery electric vehicle. The first large-scale study was performed by ADAC, Germany’s roadside assistance service [7]. While that study has been widely cited and reported on, there is another important study in Nature Energy from the University of Birmingham and the London School of Economics (LSE). The Nature study used anonmyized MOT data from 2017 through 2023. The MOT is the UK version of the US’ annual vehicle inspection.
The key findings from the ADAC study are:
EVs break down less often than ICE
EV batteries last much longer than anticipated
EVs are cheaper to run and own in every meaningful dimension [10]
The key findings from the Nature study are:
EVs age more robustly than ICE or diesel, and age more robustly than ICE or diesel when subjected to heavier usage
EV reliability is improving much more rapidly than ICE/diesel.
Of these, the most interesting to me is that the batteries last much longer than expected, that there is a strong possibility that a BEV will last much longer than the 10-15 years life than is typical of an ICE vehicle.
As per the Nature study, BEVs are also improving rapidly on reliability metrics year over year [5] This stands to reason as ICE vehicles have been on sale since 1891, whereas modern BEVs became available starting in 1996. 1996 is generous because that first EV was GM’s EV1, which was only ever available for lease. The Tesla Roadster kicked the “modern EV” movement into first gear, with the first actual consumer offering that was more than just a compliance figleaf [9].
We are now entering that window of time where BEVs are outperforming ICE vehicles in every key metric, apart from absolute range and time to refuel. On those last two, the Chinese are focusing on eliminating those advantages with announcements of five-minute charging and 500+ mile range using new battery chemistries. I think it’s reasonable to give them the benefit of the doubt — they’ve been funding metric shit-tons of fundamental research and we can assume that we’ll see production volumes of fast-charging, long-range, chemically-stable batteries, if not this year, then in the next eighteen months.
Given the trends of (a) EVs getting better, and (b) the inexorable deployment of panels and (c) the decline in price of battery storage, we now look to long-term futures.
Longer range projections with less certainty
Places like Australia will tip into the domain of massive energy surplus. In response to these surpluses, previously energy-profligate processes will be spun up to take advantage of the free electricity. [6]
Communities will begin pressing for limits to combustion engine vehicle access (because EVs are much cleaner)
Time of use pricing for electricity will become the baseline, if not universal. Over time, all utilities will mandate customer migration to TOU.
In this new world then, your car will last at least twenty years. Your car’s battery will probably outlast the car itself, with a likely second life as a grid storage asset, or worst-case, 95% recovery of the input materials.
What? Cars that routinely last 20 years or more? That’s madness!
Given the trends discussed, which we can project with certainty, a 20 year lifespan for an EV is entirely reasonable. If my Camry is fifteen years old and still perfectly serviceable, a comparable BEV is likely to last at least 20 years, but likely longer. For car makers, that kind of lifespan presents new challenges. The simplest of course, is the threat to revenue. When you double the replacement cycle (or possibly even treble it, not beyond the realm of possibility), where’s the money going to come from for new R&D? Eventually everyone’s going to have a long-lasting EV, what then?
Consider for a moment, my Camry. It was a pretty damn good car in 2010, although the CD player was looking a little long in the tooth even then. No fancy screen though, just a grey LCD. Very basic audio controls and no bluetooth. There’s an AUX jack, but phone charging is still out of a 12v cigarette lighter jack. No fancy lane-following, AEB or adaptive cruise. A passable audio system, perfectly adequate, but no satellite. All in all, a comfortable vehicle, with most of the comforts of a middle-of-the-line sedan of that time.
Now, compare that to any car off the rental lot. I can compare because, for reasons that are not important right now, I’ve driven a lot of rental cars this past year. Table stakes for even an economy-class car are a large, bright, color screen, Android Auto or Apple CarPlay, USB-C and USB-A and Bluetooth. Most cars have some kind of cell service to map where you are and show you what the speed limit is on the road you’re on. There are countless other improvements that come standard today.
If that’s the difference fifteen years makes, and our cars are going to last twenty or more, we’re going to need some kind of upgrade path that isn’t “buy a whole new car”. If there’s one thing I know for certain (beyond that EVs are going to last an awkwardly long time) it’s that we’re going to be looking at using and keeping things around longer than we have been.
The floor mat problem and the robot chauffeur dilemma.
Let’s look at things on the Camry that are starting to wear out. Nothing important has broken, so let’s assume that the mechanicals are going to make it another five years or fifty thousand miles, whichever comes first (hint: fifty thousand miles will easily come first).
One of the things that has worn out are the floor mats. Also, we’ve lost a couple of wheel trim covers. I can buy new floor mats and it would be nice to have alloy wheels, but will it make sense? In a heavily circular economy, with the focus on repair, maintenance, durability and lifespan, of course it makes sense.
Minor upgrades instead of a whole new car? That also makes sense, or at least it will make sense for EVs.
For ICE cars, well those need to head to new lives, with their materials recovered and reused. EV Car makers will have to start building vehicles that have actual upgrade paths. In 20 years, if my EV’s still good for four hundred miles with 85% of original charge capacity, I will definitely want to be able to add on a robot chauffeur without having to buy a whole new car.
That my friends, is where we’re headed.
[1] Except possibly the US, if the current withdrawal of EV support continues.
[2] Once EVs start to offer 500+ mile range and ~5 minute charging, the ICE as a format will be history.
[3] From Renew Economy in Australia: “The knockout price was a bid of $US51.59/kWh for a four hour battery (the average was $US59/kWh), which Energy Storage News says represents a 30 per cent drop from 2024 levels, and others side was a 15 per cent fall from recent record lows.” The predicted “learning path” of battery production appears to be well underway, driven by economies from production at scale.
[4] This is probable because gas station properties are desperately polluted. They’re nearly all brownfield sites, needing extensive environmental remediation. Once people no longer need gas, there aren’t that many functions that can take a gas station and re-use that land. In California, there is a December 2025 deadline for the replacement of single-walled Underground Storage Tanks (USTs). Also in CA, 60% of tanks are older than 25 years and UST manufacturers typically warranty for 30 years. This suggests that the capital expenditures of UST replacements, coupled with the decline in gas revenues due to EV adoption, for fuel station operators in CA will cause some station owners to exit the gas business. For every gas station exit, the marginal difficulty of access for ICE vehicle owners increases. The more difficult it is to refuel, the more likely that invidividual will transition to EV, thus fuelling (supercharging?) a gas-station exit spiral.
[5] Consumer Reports noted reliability problems with EVs but also pointed out that the rate of improvement with EVs is much greater than that of ICE vehicles and that PHEVs (“hybrids”) are burdened with having two engine technologies, each of which can experience difficulties.
[6] What is most likely is that new, wastefully inefficient processes will be invented purely to soak up the excess generation. The outputs of these processes will be goods that will improve society. An easy one to see coming is some kind of CO2-sucker that creates concrete construction materials.
[7] https://www.adac.de/rund-ums-fahrzeug/unfall-schaden-panne/adac-pannenstatistik/
[8] Visual Capitalist: Experts are Hilariously Bad at Forecasting Solar Installations
[9] Summary of early EV availability in the USA
| Model | Years Available | Manufacturer | Availability | Battery Type | Range (est.) | Notes |
|---|---|---|---|---|---|---|
| GM EV1 | 1996–1999 | General Motors | Lease only (US) | Lead-acid / NiMH | 60–160 mi | First purpose-built modern EV; most units later destroyed |
| Honda EV Plus | 1997–1999 | Honda | Lease only (US) | NiMH | ~80–100 mi | Civic-based; regenerative braking; early NiMH adoption |
| AC Propulsion Tzero | 1997–2003 | AC Propulsion | Limited build (US) | Lead-acid / Li-ion | ~80–100 mi | High-performance prototype; inspired Tesla drivetrain development |
| Ford Ranger EV | 1998–2002 | Ford | Fleet + limited lease | Lead-acid / NiMH | ~65–85 mi | Electric pickup for fleets; rare in public hands |
| Chevy S-10 EV | 1997–1998 | General Motors | Fleet (US only) | Lead-acid / NiMH | ~60–90 mi | Pickup EV twin to Ranger EV |
| Toyota RAV4 EV (Gen 1) | 1997–2003 | Toyota | Lease & sale (CA) | NiMH | ~95–120 mi | First electric SUV; ~1,500 units sold/leased |
| Nissan Altra EV | 1998–2002 | Nissan | Fleet lease (US/JP) | Li-ion (early tech) | ~85–120 mi | Minivan style; used by utilities and fleets |
| Nissan Prairie Joy EV | ~1996–1999 | Nissan | Japan only (fleet) | Lead-acid | ~50–60 mi | One of Nissan's earliest EV trials |