Tesla will fix its cold-weather woes, but an all-EV future is still dead on arrival.

It’s been a rough few months for electric vehicle fans. During the January cold snap, social media sites were filled with sarcasm and pictures of Teslas stranded by freezing temperatures. Lots of “dead robots out there,” one wag put it.

In mid-January, the rental car company Hertz, previously an eager early adopter of fleet electrification, announced a big sell-off of EVs that it had only recently purchased, mainly because they proved far more expensive to maintain than advertised. The same week, Ford slashed EV production, having earlier pulled back on planned battery factories. Both Ford and GM now face higher labor costs, having negotiated epic United Auto Workers pay hikes that now include previously excluded battery factories. Adding to the woes, unsold EVs are piling up on dealer lots, spurring aggressive discounting. The big sales benefit buyers but deepen the already-massive losses of manufacturers.

Finally, in the fusillade of bad news, as Fortune reports, “no one wants to buy used EVs,” leaving EV used-car values in free fall. That’s a problem for auto companies because their finance arms have been left holding the bag on fictitious residual values for leased vehicles. According to one industry executive, the situation “has the potential to destroy billions” of dollars in value for auto firms.

And now leasing has soared to over half of all EV sales, as it’s the only way to capture the federal $7,500 tax credit for most EVs. How so? By law, that credit is supposedly available only when purchasing vehicles built with materials sourced primarily in the U.S. This domestic-sourcing feature is what it took, reportedly, to get West Virginia senator Joe Machin on board to pass the all-partisan Inflation Reduction Act, because, as he surely knew, nearly all battery materials are currently foreign-made and will remain so for ages. However, the final legislation had a surreptitious exception allowing the credit for leased vehicles built with foreign materials. Evidently, the pen is mightier than the miner.

All this bad EV news, advocates claim, is merely a symptom of a nascent industry’s growing pains. There’s some truth to that, especially for the kinds of engineering issues amenable to rapid resolution. Reliability and supply chains will improve with experience and redesigns. You can bet Elon Musk has tasked his impressive engineers to improve Tesla’s cold-weather resilience to avoid future embarrassment. And America just might, one fine day, allow domestic mining to expand and to build new refineries for the minerals needed for batteries—and for everything else.

Meantime, EV boosters note, “people keep buying them.” Again, true. Last year saw record EV sales, even if outside of China it’s still a Tesla story; over half of all EVs sold in the U.S. were Teslas. Even though the trumpeted high growth-rates are an arithmetical outcome of growth from small beginnings—something one always sees in the early days of a new product—there’s no doubt that tens of millions more consumers will happily buy an EV.

What is in doubt—in fact, what won’t happen—is realizing the aspiration of an accelerating transition to an EV-dominated future. Separating aspiration from reality wouldn’t matter if this were just a debate between advocates and skeptics making private bets. This debate matters because hundreds of billions of dollars in public spending will be deployed via the misnamed Inflation Reduction Act to push EVs into markets—and because a proposed rule from the EPA, with comparable legislation in more than a dozen states, will make it impossible to buy a new car unless it’s an EV within the decade. The unprecedented magnitude of government intervention gives EV enthusiasts confidence that it will all “spur consumer demand.”

But government diktats and largesse can’t change reality. The putative EV revolution will stall out for three main reasons, and not because of “dead robots” or the other road bumps in recent news. What will happen is that we’ll run out of money, we’ll run out of copper, and car drivers will run out of patience in putting up with inconveniences. But before unbundling these truths about the practical limits of EV dominance, we have to deal with some of the myths that anchor all EV enthusiasms.

It’s received wisdom in many social media corners that “Big Oil,” worried that EVs will radically cut oil use, is somehow funding anti-EV “misinformation.” As World Economic Forum sages have declared, “rapid growth of electric vehicles (EVs) will potentially disrupt the traditional oil market.” Credit the Wall Street Journal’s Dan Neil for noting that “some of my fellow travelers suspect there must be a conspiracy to trash-talk electrification in the media, funded by Big Oil. I take a contrary view: It didn’t take a conspiracy to make EVs look bad.” Rarely have truer words been written about EVs.

EV enthusiasts at BloombergNEF claim that “EV adoption cut demand for oil by 1.8 million barrels in 2023.” At the same time, the International Energy Agency (IEA) reports that global gasoline consumption in 2023 blew past the pre-lockdown 2019 peak, even with roughly 30 million EVs on the world’s roads, up from near zero a decade ago.

Discerning analysts will note that, globally, EVs still account for barely 2 percent of all vehicles—thus, the admonition to wait. Consider, then, the case of Norway, where EVs now account for close to 25 percent of all cars. Even there, overall on-road oil consumption has remained flat instead of collapsing. Even assuming an impossibly high goal of replacing half of the world’s cars with EVs, simple arithmetic shows that doing so would eliminate only a skosh more than 10 percent of global oil demand. That’s not nothing, but it’s hardly the end of oil. The most that one can say is that EVs will moderate the growth in oil use.

But the myth that anchors the entire edifice of subsidies, mandates, and policies to force-feed EVs on everyone is that they will radically cut CO2 emissions. Again, from the IEA: “Electric vehicles are the key technology to decarbonise road transport.” The BloombergNEF team touted that, according to its calculations, EVs in 2023 avoided “122 megatons of carbon-dioxide emissions.”

Facts and context matter. The world in 2023 saw, according to NOAA, a new peak in global CO2 emissions. The claimed 122 megatons cut by EVs sounds big, but it amounts to only 0.03 percent of global emissions. For context, oil-burning war-machines in Ukraine are adding at least that much CO2 to the atmosphere yearly. Moreover, the 122-megaton figure is a calculation, not a measurement. No one really knows how much, or how little, EVs reduce global CO2 emissions.

The problem is that you can’t measure an EV’s CO2 emissions. That’s totally unlike conventional cars, where emissions are directly measurable by the quantity of gasoline used. Further, gasoline emissions are the same wherever or whenever a car is driven, or fueled, or even built. EVs obviously don’t burn gasoline, and thus those emissions are, equally obviously, avoided. But there are emissions associated with EVs, and, according to the technical literature, everything about those numbers is highly variable, requiring estimates, guesses, and assumptions about when an EV is driven, when and where it’s recharged, and especially where the materials came from to build it in the first place.

In the real world, as opposed to the realm of PowerPoint presentations, that emissions accounting is hard to nail down because it entails information not just about consumer behaviors and grid operations but also about activities in the labyrinthine global supply chains. Much about that data is proprietary or opaque, and much of it originates with Chinese industries.

The CO2 emissions arising from building an EV before it gets driven revolve around a simple fact: a typical EV battery weighs about 1,000 pounds. That half-ton battery is made from a wide range of minerals, including copper, nickel, aluminum, graphite, and lithium. Accessing those minerals requires digging up and processing some 250 tons of earth per vehicle. All that mining, processing, and refining uses hydrocarbons and emits CO2. The critical fact found in the technical literature is that those upstream emissions vary by 300 percent or more, depending on where and when materials are mined and processed. At the higher end of known ranges, upstream battery emissions can wipe out emissions avoided by not driving a gasoline car.

Every claim made about EVs reducing emissions, whether from automakers or governments, is a rough estimate at best—and sometimes an outright guess based on averages and assumptions. In every study, one finds that authors have cherry-picked a value, typically a low one. As for the future, all the variables relevant to mining and processing battery minerals point to upstream emissions rising.

Advocates respond that, whatever the emissions benefits, it will soon all be free because EVs will be easier and cheaper to buy and use. EVs, they assert, are simpler vehicles, and thus inherently cheaper to build than their gas-fueled counterparts. But EVs aren’t simpler; they’re just differently complex. The booster’s narrative claims that the transition to EVs is the equivalent of going from horse-and-buggy to the car, and thus an “inevitable shift,” in the words of Energy Secretary Jennifer Granholm. The better analogy is that an EV is the equivalent of changing a horse’s food.

Yes, conventional cars have complex thermo-mechanical systems. Engines and automatic transmissions contain hundreds of components, mated with a simple fuel tank and pump. EVs, inversely, have a simple electric motor, but the battery pack is a complex electrochemical system made from hundreds or thousands of parts, including sensors, safety systems, cooling or heating systems, and a boatload of power electronics.

One shouldn’t be surprised that the data show that building EVs entails no less labor; it just shifts it to different components and places. Tesla, the world’s biggest non-Chinese EV maker, employs about 90 people per 1,000 cars produced per year. About 80 people are employed per 1,000 conventional cars produced. Neither figure includes the upstream labor for the materials supplied to the factories.

A conventional car’s weight is 85 percent steel and iron, wherein that upstream supply chain employs less than one person per 1,000 vehicles produced. Most of an EV’s weight lies in more exotic minerals, especially aluminum and copper. That upstream supply chain employs roughly 30 people per 1,000 EVs. Nearly all that labor is offshore.

The underlying materials requirement is the single constraint that will cause the EV stall-out before other factors kick in. All the world’s mines, both currently operating and planned, can supply only a small fraction of the 700 percent to 4,000 percent increase in various minerals that will be needed to meet the wildly ambitious EV goals. The IEA estimates that we’ll need hundreds of new mega-mines to feed factories across the “transition” landscape, and that it takes 10 to 16 years to find, plan, and open a new mine.

It bears noting that buying basic materials accounts for more than half the cost of building an EV battery. That means the future price of EVs will be dominated by the future costs of those basic materials, which, in turn, depends on guesses about the future of foreign mining and minerals industries. Consider just copper, the pillar of electrification. EVs use 300 percent to 400 percent more copper than conventional cars. Industry data show that the world will need twice as much copper as it will be producing well before aspirational EV goals are reached. Unsurprisingly, one major mining CEO observed that the coming chasm between demand and supply could trigger a ten-fold copper price hike. That alone would add about $15,000 to the cost of building an EV.

This is not a question of whether planet Earth has enough copper or other minerals; nature has abundance in all domains. The issue is one of industrial infrastructures. We have no evidence that the necessary decades-long mega-investments to expand mining have begun anywhere—certainly not in the U.S. Thus far, EV boosters have waved away the minerals challenges with facile rhetoric about recycling and, in a sure sign of technological naiveté, invoking the promise of seabed mining.

And we haven’t talked about the other engineering-economic problems with accelerating the EV revolution, such as building enough chargers, expanding the electric grid, and hoping consumers will tolerate radical increases in inconveniences.

The inconvenience of EVs boils down to the reality of very long refueling times, not range. So-called fast charging isn’t fast; it takes 30 to 60 minutes, compared with five minutes to fill a gas tank. Most consumers will chafe at such long waits. And each supercharger costs about three times as much as a gasoline pump. The oft-touted $7 billion that the Biden administration is spending on fast chargers won’t come close to meeting the need; the government’s own national labs show that we’ll need more like $100 billion in superchargers if EVs reach just 10 percent of all cars.

At-home, overnight charging, using lower-cost slow chargers, currently accounts for 90 percent of all EV users, nearly all of whom have two or three cars and a garage. But only one-third of U.S. households have a garage. Of course, enthusiasts assert that charging points can be added at parking lots and roadsides. All of it will require staggering neighborhood-grid upgrades that have neither been funded nor included in the Inflation Reduction Act’s lollapalooza of spending.

And that doesn’t count eye-wateringly expensive grid upgrades needed for on-road superchargers. To be clear, this is not about energy, but about the hardware needed to deliver the energy, especially grid-scale transformers. Today, a roadside fuel station puts an electric power load (again, not energy) on the grid equal to just one 7-Eleven store. A typical EV fueling station will have the power demand of a stadium. Highways need tens of thousands of fuel stations. Making on-road refueling as convenient, simple, and cheap as the gasoline network isn’t possible with current technology.

EV boosters typically greet all these obstacles and limits with the assertion that technological progress will solve them. For some of these items, there’s some truth to that assertion. But for EVs to become ubiquitous, we’ll need quantum-leap innovations, and history shows that mandates, taxes, and subsidies aren’t how we get that kind of progress. All they do, instead, is lock in yesterday’s technologies and waste money.

None of this is to deny that there will be millions more EVs purchased, whether the government interferes or not. EVs offer interesting, useful, and even fun features for many consumers, just as do sports cars and myriad other vehicle models. But the rate of EV adoption will slow long before there’s a battery-dominated future because, again, we’ll run out of money, copper, and political tolerance for enriching other nations—especially China, where 50 percent to 90 percent of the critical materials are now produced and will be for years yet, no matter how lawmakers rewrite the sourcing regulations. And if proposed EPA rules for an EV-dominated future do become law, count on millions of very unhappy consumers, otherwise known as voters.