Many claim that the United Auto Workers strike is, at core, a reaction to an “inevitable” transition to an all-electric-car future. They say that this future involves fewer jobs because electric vehicles (EVs) are supposedly simpler machines that therefore require far less labor to build. None of this is true.

But before we puncture this myth, let’s first note that the strike, like so much in life and politics, is about money. The UAW wants a 40 percent pay hike—and you can see why, given the staggering amounts of money sloshing around the auto industry. Automakers are enjoying record profits, much of which, rather than going toward wage increases, is being used to offset the billions of dollars in losses from building EVs. And that’s the least of it—the really big money gusher is coming from the government in the form of direct and indirect subsidies for EVs and the associated infrastructures.

We are talking about trillions, not mere hundreds of billions, of dollars deployed to force a “transition” to an all-EV future that will be fueled, in the grand green plan, by wind and solar electricity. Nearly half of the subsidies in the misnamed Inflation Reduction Act (IRA) are directed at transportation “decarbonization,” and in a remarkable feat of political camouflage, some subsidies continue without limit, including for EV batteries. For perspective, NASA’s entire budget to return to the Moon is barely a few percent of the total flooding into the automobile and energy markets.

The auto industry is getting disrupted by government interventions, not technology. At no time in modern history has the government committed so much in money and mandates to a single goal as it has now to cars and how they’re fueled. This is industrial policy on a scale never before witnessed in America, outside of wartime.

The UAW wants a piece of all that. They’re right to worry about jobs, too, but not for the reasons so often espoused by EV transitionists—namely, the idea an all-EV future will require less labor.

Even the sagacious Walter Russell Mead repeated, in a recent Wall Street Journal column, the received wisdom that a “shift from internal-combustion engines to battery-powered electric vehicles threatens to reduce global employment in automotive manufacturing and make China the dominant producer.” He’s half right, but only the second half. To be fair, he’s not alone. The car cognoscenti also repeat this claim, as per Car and Driver: “The simplicity of a battery electric vehicle is analogous to that of a digital watch. It’s uncomplicated, reliable, and cheap. By that token, a car with an internal combustion engine is a bit like a mechanical timepiece: full of complicated parts that require regular maintenance.” The simplicity claim is also key to the belief that EVs will eventually be cheaper than combustion vehicles.

But this claim is a canard. The confusion stems from the failure to recognize what amounts to a complexity swap. A conventional engine is a thermo-mechanical machine, with anywhere from hundreds to a thousand parts, which is paired with a very simple fuel-storage system: a steel tank with a single-part electric pump. In an EV, by contrast, the electric motor is indeed simple, with a couple of moving parts, but it’s paired with a battery that is a half-ton electro-chemical machine with thousands of parts and welds, wiring, complex power electronics to control power flows and ensure safety, and a cooling system.

Counterintuitively, in fact, the total EV ecosystem involves more labor per vehicle, though most of the increase is found in the manufacturing supply chain. It takes some doing to unbundle the complexities, but the stakes are high, given the epic quantities of capital and the disruptions to markets and labor involved in a switch to EVs.

Let’s start with a simple, high-level sanity check: look at the data for total employment and production for the two classes of automakers. Tesla, the pure-play icon, employs 130,000 people and produced nearly 1.5 million cars last year, with 95 percent of its revenues and, arguably, employment coming from EVs. (The rest is stationary power storage and solar modules.) That works out to about 90 employees for every 1,000 cars produced annually.

Next, consider total annual U.S. car production (about 11 million) and all U.S. automotive employment (about 900,000) at both auto companies and the array of domestic parts suppliers. That yields roughly 80 people per 1,000 cars produced annually, somewhat less than Tesla. This is a more relevant comparison than juxtaposing Tesla with, say, GM alone, since conventional automakers long ago abandoned the vertical-integration strategy Tesla embraces. (Tesla produces its own electric motors and battery cells, for example.) But there’s more to the picture.

For example, let’s drill down a level to compare the labor-intensity of just the two types of drivetrains: motor-plus-battery versus engine-plus-transmission. Again, consider Tesla and in particular its “gigafactory” in Nevada, which employs about eight people per 1,000 EV drivetrains produced. By comparison, conventional engine and transmission factories require, combined, just four people per 1,000 drivetrains. That’s the inverse of the argument that EV requires less labor. (While GM’s Ultium battery factory employs just four people per 1,000 batteries, it doesn’t produce as many components as Tesla’s “battery” factory, especially the electric motors.)

There’s another complexity swap, also with relevance for labor. Conventional engines and drivetrains are controlled, in the main, by exquisitely designed mechanical linkages, valves, and pumps to manage fluid flow rates, all of which are fabricated as integral parts of the drivetrains. Meantime, controlling the transmission of massive electrical power flows in a battery-motor system requires separately manufactured and exquisitely designed high-power electronics that, ironically, cost about the same as an automatic transmission. The additional per-EV labor is hard to tease out, since the power electronics industry is dominated by Asian companies, especially China. As a proxy, we can consider that the overall semiconductor manufacturing industry entails about 1.5 employees per $1 million of sales. Automatic transmission factories employ about 1.3 people per $1 million of sales.

We can follow the labor trail further upstream to other types of industries that supply key inputs to fabricate vehicle parts, where we find still another complexity swap. About one ton of iron and steel, for instance, is used to fabricate most of the parts for a conventional vehicle. Those metals account for about 85 percent of the weight of a car. Because iron is so common (the fourth-most-abundant element in the Earth’s crust), those upstream underlying financial and labor costs are modest. Iron and steel production requires fewer than 0.5 jobs per 1,000 cars. In addition, domestic production accounts for more than three-fourths of U.S. steel consumed. (The U.S. is a net exporter of iron ore.)

The minerals picture is more complex with EVs. About one ton of rarer, more labor-intensive, and expensive minerals are used to produce each EV. An EV’s half-ton battery needs a suite of minerals ranging from copper and aluminum to graphite and magnesium (to name just a few), as well as lithium. Each EV also uses, relative to a gas-powered car, about a quarter ton more copper to build electric motors and for power transmission and about a quarter ton more aluminum (instead of steel) to make the body and frame lighter to offset the battery’s weight penalty. Obtaining that total of about one ton of refined minerals involves labor to dig up, move, process, and refine about 500 tons of the Earth per car. (Yes, digging is also required for iron and steel, but less than one-tenth as much, because iron is so common.)

For perspective on the hidden labor involved, think about copper, where the data show that mining and refining collectively employ roughly eight people for every 1,000 EVs. Add in the other minerals, and the total upstream labor looks like roughly 30 people for every 1,000 EVs produced—a 60-fold increase compared with the mineral supply chain for a conventional vehicle. Note that there aren’t any comprehensive analyses of all these various labor features in the two vehicle ecosystems; there should be. The numbers we’re using here are estimates, based on public data from companies in the relevant industries.

As Walter Russell Mead correctly noted, China utterly dominates the refining of EV minerals and Chinese global investments dominate much of the mining of them. If one were to say that “lithium is the new oil,” then China would be the new OPEC. Let’s credit Elon Musk with expanding Tesla’s vertical integration by breaking ground recently to build a $1 billion lithium refinery in Corpus Christi, Texas. But no amount of subsidies will allow America, in the near future, to refine and fabricate most EV components and inputs—and particularly not the necessary underlying raw and refined minerals. Nevertheless, Inflation Reduction Act subsidies appear to be flowing to mining companies in Australia, for example. And if the Biden administration has its way, some of those subsidies could even flow to Saudi Arabia, which is being encouraged to build minerals refineries supplied by expanding mining in Africa.

As for the claims of simplicity when it comes to fueling EVs, it’s obvious that on-the-road fast-charging will be critical to realizing the vision for an all-EV future. In that regard, it’s also worth noting that one needs roughly the same number of on-road fuel-stations as conventional cars already have, but the labor associated with delivering energy to those EV fueling stations is greater per 1,000 cars than it is for gasoline. 

The EV transitionists have at least one thing right: this ostensibly “green” path will create more jobs, even if many of those jobs are elsewhere. But, in economic terms, more labor-per-car is inherently inflationary. Engineers have spent the past century trying to reduce labor (and costs) per vehicle. Now policymakers are eager to throw away that progress in pursuit of reductions in carbon dioxide emissions.

And this brings us to a final illusion of simplicity, one associated with accounting for the carbon dioxide produced by EVs. Carbon emissions from a conventional car are easily documented and precisely known; one just counts the gallons of gasoline bought. But carbon-counting with EVs is not so simple.

To know the actual CO2 emissions associated with driving an EV, one must know precisely when and where each kilowatt-hour is put into a battery. Those emissions can vary from near zero to as much as burning gasoline for the same miles. But that accounts for only half of the complexity. It is, for all practical purposes, impossible to know how much CO2 emissions are produced from building the vehicle, caused by the energy used to dig up and process the 500 tons of earth needed to fabricate each EV. We do know that those emissions can be large enough to wipe out much, perhaps all, of the emissions avoided by not burning gasoline in the first place. (More details on all this can be found in my recent Manhattan Institute report.)

The bottom line, by now, should be obvious. A monomaniacal attempt to create an all-EV future, especially in the time frames envisioned, involves not only more overall labor but an unprecedented offshoring of labor, as well as a massive misallocation of capital. The ultimate result will be economic havoc and bankruptcies—and that will certainly lead to fewer jobs.

A summer 2023 Pew Research Center survey found that a rising share of Americans—now six out of ten—oppose “the idea of phasing out the production of new gas-powered vehicles by 2035.” That share will likely rise as more people, perhaps unavoidably, learn about the realities of vehicle supply chains.

Photo by JEFF KOWALSKY/AFP via Getty Images


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next