On April 17, the premier journal NeuroImage’s entire editorial team, comprising more than 40 scientists, resigned over the “unethical fees” charged by the journal’s academic publisher, Elsevier. With more than $2 billion in annual revenue, the publisher’s profit margin approaches 40 percent—rivaling that of Apple and Google. “Elsevier has become kind of like the poster child for evil publishing companies,” said neuroscientist Kristen Kennedy, one of the recently resigned senior editors.

Kennedy relies on taxpayer money to study the aging brain. At the University of Texas at Dallas, federal grants help fund the staff, equipment, and experiments in her lab. But this public money, largely from the National Institutes of Health, is being drained by exorbitant publishing fees.

When scientists have a discovery worth sharing, they seek publication in scholarly journals. This “publish or perish” cycle has been trusted for centuries, as scientists rely on publishing to attract grants, maintain their jobs, and bolster their reputations. NeuroImage, a top journal in Kennedy’s field, charges scientists $3,450 to publish a single paper. When a typical productive lab publishes eight to 12 papers a year, the costs become considerable. Today, scientific publishing has become a multibillion-dollar business that not everyone can afford.

“Elsevier continues to prey on the academic community, claiming huge profits while adding little value to science,” wrote Chris Chambers, a professor at Cardiff University and fellow former editor.

The NeuroImage editors resigned after Elsevier refused to lower the “unethical and unsustainable” publication fees from $3,450 to under $2,000. Though Elsevier is not transparent about its budget, Shella Keilholz, a former NeuroImage senior editor, said the editors estimated that direct article costs—such as paying for submission and handling software, proofing and typesetting papers, and website hosting—were less than $1,000. Major corporations like Elsevier hold tremendous power over the reputations and reach of scientists—and they’re making billions from it.

“From Elsevier’s standpoint, it’s just a profit-making machine,” said Keilholz, a professor of biomedical engineering at Georgia Tech and Emory University. “The question is whether it’s ethical to do that. And in our community, we think not. We think that public money is being wasted.”

These journals run on volunteer labor: All peer reviews are done for free, and NeuroImage’s stipends for editors are a mere $1,500 to $2,000 a year. To Keilholz, it’s hard to understand what the authors are paying for. In recent years, the NeuroImage editors received an increasing number of complaints from scientists—especially early-career researchers and those from low- and middle-income countries—that the fees were too high. While Elsevier provides certain waivers, many scientists are left behind. “It’s creating a two-tiered system where only the rich labs can afford to publish in those journals at all,” said Lucina Uddin, a former NeuroImage editor and professor at University of California, Los Angeles. In 2012, scientists called for a boycott of Elsevier, amassing more than 20,000 signatures from academics refusing to support “any Elsevier journal unless they radically change how they operate.” A 2016 study found that 38 percent of signatories eventually abandoned their commitment. Since then, the fees have only gotten higher.

“We value very highly our editors and are disappointed with the decision of the NeuroImage Editorial Board to step down from their roles, especially as we have been engaging constructively with them over the last couple of years,” according to a statement provided in an email by an Elsevier spokesperson. Elsevier argued that its fee is “competitively below the market average relative to quality.” For instance, Springer, another major publisher, charges a publication fee of $11,690 for its journal Nature Neuroscience, while fellow publishing giant Wiley charges $3,850 for its journal Human Brain Mapping.

According to Kennedy, Elsevier told editors that fees were based on a journal’s reputation—specifically, their “impact factor.” As the editors grew the journal’s prestige, Elsevier increased the publication fee by about 15 percent, jumping from $3,000 in January 2020 to $3,450 in 2021. It was inevitable, Keilholz felt, that the fees would keep rising. She concluded that the incentives for for-profit publishers were not aligned with “what we want for science.”

So the entire editorial team left Elsevier.

“We’re creating the science. We’re writing the papers. We’re editing them and we’re reviewing them—we’re the entire workflow. So we took the power back by realizing we’re the ones who need to be satisfied by this,” said Uddin. “We can just say negotiation time is over.”

Working with the nonprofit publisher MIT Press, the editors are launching their own nonprofit, open-access journal, called Imaging Neuroscience. At $1,600, their lower publication fees will cover direct article costs and the waived payments for researchers from lower-income countries, explained Keilholz. The editors have chosen to forgo stipends until they hit a rate of 500 papers a year. More than a thousand scientists have already signed up to review for them, and 20 papers were received in the first week. Their goal is to “out-prestige” their former employer.

On the day the resignation was announced, Michael Demidenko, a postdoctoral researcher at Stanford University, had planned to submit a report to a companion journal of NeuroImage. Now, he’s pivoting to Imaging Neuroscience. Still, he says it’s a risk for early-career scientists. You have to hope that hiring committees share your viewpoint about “open science and against exorbitant fees feeding this billion-dollar corporation,” said Demidenko. If not, the lack of a premier journal’s name on your résumé could lose you a rare spot in academia.

In recent years, journals have moved away from the subscription model—in which scientific articles are pay-to-read—to an open-access format. This move could potentially make science more accessible and equitable not just to other scientists but to everyone. “It’s silly for governments and taxpayers to fund research that they can’t afford to see. It’s unfair and it’s not good for society,” said MacKenzie Smith, the University Librarian for University of California, Davis.

But as paywalls drop, scientists are the ones bearing the brunt of the cost. Before open access, journals profited from subscription fees. Under this system, readers either subscribed to entire journals or purchased individual articles. According to 2018 research by Duke Libraries, the average cost of a paywalled Nature article was $32. The study offered this calculation: If the 244,133 authors who cited the popular article “Cleavage of Structural Proteins” paid out-of-pocket to access it, the total revenue would be $7,821,256.

“When we made the switch to open access, publishing companies had to find another way to make these extremely crazy profits that they’d gotten used to,” Keilholz said. “So they started charging article processing fees.” Before NeuroImage became fully open access in January 2020, scientists didn’t pay to publish. Immediately after, the publication began charging $3,000.

David Liberles, professor at Temple University and editor in chief of the Springer-owned Journal of Molecular Evolution (which has a publication fee of $3,690), has pushed for his journal to remain hybrid. Scientists can either pick the subscription model and publish for free, or pay the fee to publish for open access. “I think the purely open-access model is broken,” said Liberles. He is concerned that people are discouraged from publishing novel ideas in the literature when it means coughing up thousands of dollars.

According to Jeff MacKie-Mason, the UC Berkeley University librarian, there was worry among scientists that a “pay to publish” system would incentivize publishers to accept anything as long as someone writes a check. He has been tracking issues among for-profit publishers, including a trend of editorial boards feeling pressured to lower scientific standards in favor of increasing publisher profits. For example, fellow publishing giant Wiley recently came under scrutiny for pressuring the editors of The Journal of Political Philosophy to publish more articles a year, despite editors’ concern about quality. MacKie-Mason concluded that academics are pushing back. In March, more than 50 journals—including some owned by Elsevier and Springer—were delisted by the global research database Web of Science because of failing quality standards. Meanwhile, hundreds of articles have been retracted. “We’re trying to create a legacy with the science we create,” he said, and that means scientists must discipline profit-minded publishers.

However, the top publishers are growing more powerful. A handful of companies control most scientific journals after the last two decades saw a major wave of scientific publisher mergers and acquisitions. One study found that more than half of all scientific articles published since 2006 have been controlled by just five companies: Elsevier, Taylor & Francis, Wiley-Blackwell, Springer, and Sage. In 1996, these publishing houses owned only 30 percent of the journals. Now it’s hard to avoid them. Tenure committees and funding agencies want to see people publish in high-impact journals to show that “your work is important,” Keilholz explained. And the more the scientific community buys into this system, the more publishing monopolies can raise their fees.

But MacKie-Mason, who is also a professor of economics, sees a way out of this cycle. “No monopolist can charge an infinite price.” MacKie-Mason called on academia to push back against exploitative prices, through dramatic gestures such as the NeuroImage editorial board’s. When the editorial team says “something’s rotten here and the publisher is misbehaving,” it’s up to them to “police the industry.”

The former NeuroImage editors hope to ditch the for-profit system entirely. If they can grow Imaging Neuroscience to have NeuroImage’s impact, other scientists may be inspired to follow. Smith has already seen an uptick in the number of editors at UC Davis “looking for a way out.” She’s helping them brainstorm about where they could move their journal to, or potentially start a new one. But there aren’t enough nonprofit publishers to support every journal. MIT Press currently publishes 14 open-access journals and has the capacity to open only one or two new ones a year. “I don’t believe that all of these big publishers and their journals are going to disappear. So in the short term, we really need to just get better at forcing them to do better,” Smith said. “But we’ll keep working on alternatives that may someday take over.”

As the academic publishing industry balloons into a $25 billion business, its monopolization may threaten how far public money can go toward funding scientific breakthroughs. Whether pushing for open access or subscription, for-profit or non-profit, these scientists want a change. “That money is instead just going to the publishing companies, and not really accomplishing anything,” said Keilholz. “That’s not moving science forward. It’s basically an attack on science.”