There’s Surprisingly Little Evidence That EVs Will Require Fewer Workers

When former President Donald Trump was campaigning in Michigan last week, he warned autoworkers that President Biden’s electric vehicle policies would “put an end” to their “way of life.”

“Hundreds of thousands of American jobs, your jobs, will be gone forever,” he said. “By most estimates, under Biden’s electric vehicle mandate, 40% of all U.S. auto jobs will disappear.”

Trump may be exaggerating, but the underlying idea, that electric vehicles require less labor to manufacture than internal combustion engine cars, is the conventional wisdom. It has been circulated for years by automakers, autoworkers, politicians, and journalists. EVs contain fewer parts, the thinking goes, so naturally they will require fewer workers.

That logic seems obvious, which might be why it hasn’t received much scrutiny. But when I tried to find any research supporting it, what I found instead suggested the opposite. A number of analyses showed that electric vehicles could actually require more labor to build than gas-powered cars in the U.S., at least for the foreseeable future.

There are countless news articles and studies that reiterate the point that electric vehicles “have fewer moving parts” or are “less complex” and therefore pose a threat to autoworkers’ jobs. Many cite a 2017 Ford presentation that mentioned a “30% reduction in hours per unit” as a benefit of producing EVs, or former Volkswagen CEO Herbert Diess, who said in 2019 the company would need to make job cuts due to its switch to EVs, which “involve some 30% less effort.” More recently, as the United Auto Workers strike has ramped up, a 2022 quote from Ford’s CEO Jim Farley that “it takes 40% less labor to make an electric car,” has been circulating.

But I couldn’t find any data, research, or even further explanation backing up these figures. Part of the challenge of digging into these claims is that it’s not clear what they even refer to. Are the CEOs talking about the labor required for final assembly, like dropping in the motor and putting on the doors? Are they taking into account the production of components, like the EV battery? Where do they draw the line on what constitutes EV manufacturing?

Get one great climate story in your inbox every day:

Ford didn’t respond directly to my request for more information about its public estimates. Instead, spokesperson Dan Barbossa replied that if I was going to quote Farley, I needed to include his entire quote. After dropping the “40% less labor” statistic, Farley had continued, “So as a family company, we have to insource so that everyone has a role in this world. We have a whole new supply chain to fill out, in batteries and motors and electronics.”

There may be more to Farley’s words than a bit of public relations fluff. His suggestion that building out new supply chains will help people find “a role” aligns with the conclusions of a study that Volkswagen’s independent Sustainability Council commissioned in 2020. It was conducted by the Fraunhofer Institute for Industrial Engineering, a German research group, using Volkswagen company data, and found only minor impacts on employment due to the transition. Losses can be mitigated by “shifting to the production of new components,” it said, like the individual battery cells that make up the battery packs.

One of the findings was that “employment intensity” for the final manufacturing of Volkswagen’s electric ID.3 is only 3% lower than that of the conventional Golf Mk8. The bigger gap is in the labor required to produce the individual components of each car’s drivetrain. The employment intensity of the battery system and electric motor, combined, was about 40% lower than that of the combustion engine and transmission system.

Notably, the study did not include the jobs required to produce the individual battery cells which make up the battery system, because Volkswagen wasn’t producing them at the time. But a more recent analysis of the U.S. manufacturing landscape found that cell production holds the most potential for job creation, and concluded that if you account for this, the transition to EVs could actually result in significantly more jobs.

Turner Cotterman, a McKinsey consultant, led the research as part of his Ph.D. in public policy and engineering at Carnegie Mellon under Associate Professor Kate Whitefoot. He sought out partnerships with U.S.-based automakers and electric vehicle component manufacturers and collected original data from nine companies on the number of hours it takes to complete more than 250 process steps. In some cases he visited the shop floors and personally gathered the data himself. In his final analysis, he also incorporated public data for an additional 78 production process steps. He used the data to model three scenarios where EV and combustion engine powertrains are produced at the average efficiency, as well as a “most efficient” case and a “least efficient” case.

In every case, EV manufacturing required more hours. The conventional powertrains took 4 to 11 worker hours, while the EV powertrains took 15 to 24. “A lot of the confusion sits around, what parts are you counting in this evaluation?” Cotterman told me. “We’re saying that if you were to produce every single component in an EV in the U.S., that the total sum of those powertrain components will be higher than the equivalent ICE components.”

Cotterman, Turner and Fuchs, Erica Renee and Whitefoot, Kate, The transition to electrified vehicles: Evaluating the labor demand of manufacturing conventional versus battery electric vehicle powertrains (June 4, 2022)

There are a few important caveats to the research. For one, Cotterman stressed that these are present-day numbers, and they might change as EV plants scale up and learn to be more efficient. When he looked at data from Chinese manufacturing plants, they were a lot more efficient than what he saw in the U.S. And that relates to his other point. Currently, most battery components are not made in the U.S.

“With so many battery components made in China and South Korea, a lot of those potential labor hours are being captured by other countries,” he said. “So it's a question of the future American manufacturing workforce — how do we value them? How many opportunities do we want to extend to them?”

Another report published in 2021 by the Economic Policy Institute, a nonpartisan think tank, reached a similar conclusion. It found that the stakes for workers in the EV transition depend largely on public policy efforts to shore up U.S. manufacturing and enhance job quality. “The real challenge is making sure U.S.-based producers can invest enough to become competitive in battery production, and claw back some of the overall sales market share they lost since the Great Recession,” Josh Bivens, chief economist at the institute, told me in an email. “These are much bigger deals than anything about the inherent production process of EVs — and they’re very amenable to policy.”

Automakers have claimed that paying workers more would put them at a disadvantage and hinder their ability to invest in the EV transition. But in a recent blog post, the Economic Policy Institute argued that with the help of subsidies from President Biden’s signature climate law, the Inflation Reduction Act, automakers have “more than enough money” to invest in EVs, pay workers a fair share, and maintain healthy profits.

The IRA created a domestic manufacturing tax credit that subsidizes the production of battery cells to the tune of $35 per kilowatt-hour of capacity. It offers an additional $10 per kilowatt-hour tax credit for the domestic production of battery modules, or the process of assembling the cells into arrays that later get put into battery packs. And there’s another incentive for automakers to onshore battery production — it will help their vehicles qualify for the IRA’s consumer tax credit.

According to a database maintained by the advocacy group Climate Power, there have been about 10 EV battery manufacturing plant projects announced in the U.S. since the IRA was passed, at least some of which will produce cells.

So is the crux of the matter that EV job losses or gains all come down to batteries? Not necessarily.

Whether or not the U.S. is able to build up domestic battery production, early evidence of the EV transition in the United States shows that EVs may require more labor, even in the final assembly stages.

Anna Stefanopoulou, a professor of mechanical engineering at the University of Michigan, has been investigating three manufacturing sites that used to produce conventional cars and are now producing EVs: A Tesla factory in California that used to be a jointly-owned facility between GM and Toyota that produced Pontiacs and Corollas; a Rivian plant in Illinois that previously produced Mitsubishis; and the Orion Assembly plant in Michigan, where GM transitioned from producing Chevy Sonics and Buick Veranos to electric Chevy Bolts.

Her research has not been peer reviewed or published yet, but Stefanopoulou told me that after analyzing publicly available data sources for employment and output at each plant, she found that productivity had gone down in all three cases. Each one is producing fewer vehicles per worker than they were before, meaning it’s taking more people per vehicle to produce electric cars. The California site, which has been producing EVs for the longest out of the three, showed the most dramatic change. At its peak, the GM/Toyota plant produced 80 vehicles per person per year. The Tesla plant averages 30.

Stefanopoulou believes the data reflects the nascent state of U.S. electric vehicle manufacturing. She predicts that after a decade or so, as processes become more streamlined, the commonly-held belief that EV assembly requires less labor will turn out to be correct. However, she also said that if she were to consider battery cell production, as Cotterman did, EV production on the whole could require more people.

She also stressed that her data is not conclusive, and poses many more questions. For example, she found that overall production per worker in the U.S. is falling. So does the labor intensity at the EV plants reflect something specific about those factories, or a bigger issue in U.S. manufacturing productivity?

It’s also been hard for her team to identify what was actually being produced at each plant at any given time. For example, the previous owners of the California plant did not assemble engines there, but the Tesla factory is assembling battery packs. So that might explain why productivity is so much lower now. But there are a lot of unknowns. “Over the years, they changed their patterns,” she told me. “They take the cells and assemble the pack, or occasionally they manufacture cells. So we don’t know exactly what kind of work the plants include. We know the outputs are vehicles, but what does assembly include?”

In any case, Stefanopoulou is torn about what conclusion to draw from her findings on productivity. “Sometimes I don’t know if what I will present in my paper will be good news or bad news,” she told me. “Maybe it’s good news for our people that are involved, but at the end, you know, we need to be productive also, so that we can actually lower the costs so people can afford buying electric vehicles.”

What seems clear is that whether the transition results in more jobs or fewer depends a lot on which processes you’re including, how many of them will ultimately be done domestically, and how much will get streamlined through automation and other efficiency measures.

At the same time, topline job numbers aren’t the full story. The jobs created in the EV transition will certainly not all resemble the jobs that are lost. They may not be located in the same places, or require the same set of skills. Workers are right to be worried about upheaval.

But these are things that can be managed, if automakers are willing to come to the table with workers, and vice versa. For example, when Ford negotiated the closure of its Romeo Engine Plant at the end of last year, every employee was offered either a buyout or a transfer to another facility. Barbossa, the Ford spokesperson, told me many are now working about 20 minutes away, at the Van Dyke Electric Powertrain Center, building EV power units for the F-150 Lightning and hybrid powertrains for the Maverick and F-150.

I reached out to the United Autoworkers to get their thoughts on these studies, but the union did not respond to my questions. The UAW does appear to have a good handle on the stakes of battery manufacturing, however. Last week, Jim Farley of Ford provided an update on the negotiations, and said that “the UAW is holding the deal hostage over the battery plants.”

Farley vowed that none of its workers will lose their jobs due to battery plants during the next contract period. “In fact, for the foreseeable future we will have to hire more workers as some workers retire, in order to keep up with demand,” he said. “We are open to working with the union on a fair deal for battery plants, but these are multi-billion investments and they have to make business sense.”

Read more about electric vehicles and labor:

What the UAW Wants Exactly — and What It Means for Electric Cars