A Revolution in Science Publishing, or Business as Usual?

Each year, governments around the world pour vast sums of public money into scientific research — as much as $156 billion in the United States alone. Scientists then use that funding to further human understanding of the world, and occasionally to make compelling discoveries about everything from whale brains to dwarf stars to the genetic underpinnings of deadly cancers. But often, this research — despite being subsidized with taxpayer money — ends up being published in exclusive journals that sit behind steep paywalls with three- and four-figure subscription fees, accessible to only a tiny fraction of the public.

The power of these scientific publishers — with names even lay readers might recognize: Springer Nature, Taylor & Francis, Elsevier, among a handful of others — is substantial. According to one estimate, just four corporations now publish close to 50 percent of scientific papers. Together, they control the copyright to much of the world’s scientific literature, charging billions of dollars each year for access to that body of knowledge — and securing hefty profits in the process.

“If knowledge is a human right, why can’t we manage it as a commons, in collaborative ways managed by the academic community, not by for-profit initiatives?”

Critics have argued for decades that this system is wasteful, and that the public should have access to the scientific literature that its tax dollars support. Scientists, scholars, and public institutions, they say — and not the private sector — should control access to this trove of knowledge. “The commercial interests of publishers trying to promote their brand should not be what determines what kind of scientific discipline becomes well-funded and well populated,” said Michael Eisen, a biologist at the University of California, Berkeley and a vocal supporter of the alternative model of research distribution, broadly referred to as “open-access” publishing, which has long aimed to harness the internet to make research more widely available — at little to no cost. The current system, he said, gives a handful of publishers “a disproportionate power to shape the way that science is done.”

The ensuing decades have been, in certain respects, a triumph for supporters of open access. Research funders in the United States and Europe adopted policies to make more of the research they fund accessible to the public. Several open-access organizations now operate thriving journals, and pirating tools like Sci-Hub have made it easier than ever to sneak around publishers’ paywalls.

Meanwhile, a group of Europe’s largest scientific backers — including the funding agencies of France, Britain, and the European Union as a whole — will soon require all research they underwrite to be openly accessible to everyone. That scheme, called Plan S, may be the most ambitious government-sponsored open-access effort yet — though federal officials in the U.S. are considering a policy that would require immediate open-access publishing for all federally-funded research as well, potentially revolutionizing the publishing industry. “Open Access Is Going Mainstream,” The Chronicle of Higher Education announced in a headline last year.

These successes, though, have also revealed divisions within the open-access community over two now-familiar questions: Who should run the publishing houses? And who should pay for the whole system? Instead of an open-access commons run by scholars in the public interest, the new open-access revolution increasingly looks like it will depend on the same big commercial publishers, who, rather than charging subscription prices to readers, are now flipping the model and charging researchers a fee to publish their work. The result is a kind of commercial open-access — a model very different than what many open-access activists envisioned.

Some advocates see corporate open-access as a pragmatic way of opening up research to the masses. But others see the new model as a corruption of the original vision — one that will continue to funnel billions of dollars into big publishing companies, marginalize scientists in lower income countries, and fail to fix deeper, systemic problems in scientific publishing.

As it stands, all trends point to an open-access future. The question now is what kind of open-access model it will be — and what that future may mean for the way new science gets evaluated, published, and shared. “We don’t know why we should accept that open access is a market,” said Dominique Babini, the open-access adviser to the Latin American Council of Social Sciences and a prominent critic of commercial open-access models. “If knowledge is a human right, why can’t we manage it as a commons, in collaborative ways managed by the academic community, not by for-profit initiatives?”


At the center of the debate is a process of peer review and publication with roots in the 18th century. Under this model, when you submit a paper to a journal, it goes to a group of editors, who are sometimes academics receiving little or no pay for this additional editorial work. If the editors think the paper seems promising, they shop it out to peer reviewers, who read the draft, recommend whether to accept it, and offer feedback, sometimes over the course of multiple rounds of revisions. Those peer reviewers typically don’t get paid for their work. Instead, they volunteer their time as a service to the field, following centuries-old conventions of scholarly collaboration.

Eventually, the editors bundle those individual articles together into an issue of the journal and send it out into the world.

The journal’s publisher does not pay for any of that research, writing, or review. But it still has to coordinate the whole review process, edit the final product, and print and distribute physical copies of the journal. Publishers also assume the responsibility for the content in their journals. To cover those costs, they charge subscriptions to individuals and university libraries, using complex, sometimes opaque methods to negotiate and set price packages with particular institutions.

“We do get bothered when we see certain publishers getting money in their pockets as a result of the stuff we’re doing for the good of our community.”

Before World War II, university presses and learned societies handled most of that labor, often covering their expenses with funding from wealthy patrons, universities, and academic organizations, or from the article’s own authors. Their chief goal was to circulate new research rather than make a profit, and they typically published no more than a few journal titles. The arrangement began to change after the war. Eager to build on wartime scientific advances and keep pace with geopolitical rivals, governments infused more cash into scientific research. The budget of the National Institutes of Health, a major U.S. government funder, increased nearly 20-fold between 1945 and 1950 — and then kept growing.

In the 1950s, a group of for-profit publishers began publishing more academic journals. They had capabilities that the old university presses and societies did not. They were nimble, able to identify new fields and start new journals. They could market their journals around the world, fueling the globalization of academic research. And they could keep down costs by developing economies of scale.

By acquiring old journals, starting new ones, and entering revenue-sharing agreements with some academic societies, the most successful publishing houses, like Pergamon Press and Elsevier, eventually came to own or publish hundreds of titles. Commercial publishers “were entrepreneurial, saw the gaps in the market, created journals, and therefore really helped those fields develop,” said Aileen Fyfe, an historian at the University of St. Andrews in Scotland, who studies scholarly publishing.

The new commercial publishers also adopted a similar system of peer review. Academics continued to contribute free labor to the process, as editors and reviewers. What had changed, of course, was that much of that work was now feeding corporate profits — in effect, a massive contribution of free academic labor to a highly lucrative business model.

Tensions began to rise in the 1980s and 1990s, as university library budgets shrank. At the same time, commercial publishers began to merge. Big publishers started ratcheting up prices, placing a crunch on academic libraries. Indeed, between 1986 and 2004, according to one estimate, subscription prices rose by 273 percent, much faster than inflation.

As subscription prices rose, those labor-for-free arrangements began to seem less tenable to some scholars. It was in the early 2000s, Fyfe estimates, that some academics began to complain that their labor was feeding an unfair system — a system that, in effect, took research for free, packaged it, and then sold it back to the very institutions that had paid to produce it, while yielding profit margins that, advocates complained, sometimes approached 35 percent.

With these realities in mind, more than a dozen scholars and librarians issued a manifesto in 2002 — sometimes called the Budapest Declaration — that laid down the principles for the open-access model of publishing, in which all papers would be free for anyone to read. The Budapest Declaration makes two simple observations: While traditional publishers have long argued that vetting, publishing, and distributing scientific work is expensive business, the internet has made it easy and cheap to share that knowledge. As a result, the group reasoned, it should be possible to circulate that scientific and academic research without spending nearly so much money. They imagined a world where scholars controlled their own publishing, putting their work online where it would be available, in perpetuity, for free.

The declaration was and remains a little fuzzy on the details of how, exactly, that new system would pay for itself. Perhaps government and foundation grants would cover the costs? Or universities would chip in for those funds? Regardless, advocates have argued, an open-access world would be worth the cost. Indeed, it would lead to better science, by offering more people access to research, and by freeing scientific publishing from the pressures of the marketplace.

“If we were doing this work for the sake of the discipline, and it was all in a nonprofit kind of world, I don’t think we’d be so bothered by it,” Fyfe said. “But we do get bothered when we see certain publishers getting money in their pockets as a result of the stuff we’re doing for the good of our community.”


In fact, early open-access advocates in Europe and North America were gesturing toward a system that already existed. In Latin America, universities, academic consortia, and scholarly organizations have always run most of their own publishing, with the online material available to anyone for free. The publication costs per article, advocates say, are a fraction of those of commercial publishers. And those costs are typically underwritten by university budgets and direct government funding.

Written in Spanish and Portuguese, Latin American journals have not historically attracted the interest of commercial publishers. Instead, university-based presses published academic journals as part of their broader, state-funded educational missions. Often, rather than charging subscriptions, these presses swapped titles with other institutions, forming an informal, region-wide system of barter and exchange that, as open-access scholar Juan Pablo Alperin described it, functioned as a kind of “pre-incarnation of open access.”

Got questions or thoughts to share on Covid-19?
Email Undark at [email protected].

As Latin American journals began to transition online in the late 1990s and 2000s, they brought that knowledge-sharing model into the digital world. That system that remains in place today. “Our colleagues in Mexico, in Latin American countries — they are way ahead than a lot of the publishing platforms in Europe because they are not stuck in a certain legacy system,” said Leslie Chan, a development studies scholar at the University of Toronto-Scarborough and a signatory of the Budapest Declaration.

That economic model reflects deep-seated ideals about the nature of academic inquiry. “For us, knowledge is a human right,” said Babini. Open access allows all scholars to participate in the conversation without the burden of publication fees.

It was perhaps once possible to imagine that research in Europe and North America would follow this route. Starting in the 1990s and early 2000s, some scientists began starting their own journals and publishing houses. The best known of these, the Public Library of Science, or PLOS, launched its first journal in 2003. Today, it offers several online-only, open-access journals, publishing tens of thousands of articles per year.

For-profit outfits like Ubiquity Press, launched by a trained archaeologist in the U.K. in 2012, also emerged to offer low-cost, open-access options, most of them digital-only. And some university presses in the U.S. began hosting open-access journals.

“At the end of the day, they produce a PDF on a webpage and they archive it. That’s basically it. And it costs the same as a car to do that!”

Analysts predicted that this change would devastate commercial publishers. For example, in 2012, as government funders pushed for more open-access offerings, the London-based industry analysts Claudio Aspesi, Andrea Rosso, and Richard Wielechowski warned investors that the future “would be catastrophic” for Reed Elsevier, now known as RELX Group, which houses Elsevier, the world’s largest publisher of academic journals.

Instead, the transition to open access has gone in a different direction — one increasingly dominated by the very commercial publishers that early open-access advocates once hoped to marginalize. These publishers have embraced something called the article processing charge, or APC — a fee charged to the paper’s authors in exchange for making their work open access. The APC-centered system allows publishers to make money on the front end (by charging the writer) rather than the back end (by charging the reader).

Smaller open-access presses, including PLOS and Ubiquity, charge APCs. But in the past several years, commercial publishers have entered the APC-funded world with gusto. The main model they use is the hybrid journal, in which some articles are open access, and some are paywalled. The publisher can still sell subscriptions to the journal as a whole, plus they can collect APCs on individual open-access articles.

And those APCs are often lucrative. Costs vary widely but generally top $1,500 per article. In prestigious journals, the cost of an open-access article can be more than $5,000. Nature Communications, a prominent journal, charges an APC of $5,380.

Those kinds of rates infuriate many open-access advocates. Commenting on the Nature Communications APC, paleontologist and longtime open-access advocate Jonathan Tennant acknowledged that the journal’s publisher, Springer Nature, has to do the work of sifting through submissions, rejecting papers, and lining up peer reviewers. “But at the end of the day, they produce a PDF on a webpage and they archive it,” he said. “That’s basically it. And it costs the same as a car to do that!”


Rachel Scheer, a spokesperson for Springer Nature, declined to provide a breakdown of how Springer Nature allocates its APC funds. Tom Reller, as vice president of communications for Elsevier, also declined to share a breakdown of APC distribution for Elsevier journals. But, in a detailed email to Undark, he wrote that the average APC for Elsevier’s hybrid journals is $2,586. (The average is lower for Elsevier’s purely open-access titles).

Reller, who left Elsevier earlier this year, wrote that open access had been “really positive” for the company’s business model. “We fully support open access and are one of the largest open-access publishers in the world,” he added. Reller also challenged common estimates of profit margins — and, he wrote, “whether or not our margin is 37 percent, or 3.7 percent, we’re still charging a competitive rate in the market, and in fact providing more volume, at higher quality and at a lower price per unit each year.”

“Whether you like it or not, the publishers are there, and, whether you like it or not, that’s where the researchers like to publish.”

Just two years later, Aspesi, the analyst who predicted catastrophe for Reed Elsevier, had reversed his judgment. In a 2014 report, he and a colleague, Helen Luong even suggested that “OA funding may in fact be adding to the profits” of commercial scientific publishers.

Some open-access advocates see this as a valuable compromise. Johan Rooryck, a linguist at Leiden University in the Netherlands and the official “open-access champion” for Plan S, said that, ideally, universities will take the money they spend on subscriptions, and gradually flip that cash toward hybrid deals that pay for both subscriptions and APCs — an arrangement called a transformative agreement. “We are really aiming for cost-neutral transformative agreements,” he said. Under these agreements, universities continue to pay large sums to commercial publishers, but, over time, more and more of the research they produce will be available to the public for free.

Rooryck acknowledges that such a scenario will not satisfy advocates who envision the wholesale dismantling of commercial publishers. “I have a lot of sympathy for those people,” Rooryck said, praising the work of non-commercial presses. “If we were to redesign academic publishing from scratch, that’s how we should do it,” he said. “But we are not working from scratch. We are working from the reality of today, and, whether you like it or not, the publishers are there, and, whether you like it or not, that’s where the researchers like to publish. So what you have to do is see how you can convert the system to a system that is fair enough.”

Tennant praised Rooryck and his colleagues, but he worries that Plan S will “maintain the present state of the industry without any financial disruption at all, without any disruption to the ways in which research is evaluated.” (After Undark’s interview with Tennant, the open-science organization OpenCon announced that it was expelling him for unspecified violations of its code of conduct. Tennant has both apologized for his actions and contested the decision in posts on his blog.)

The stakes can feel especially high for those in Latin America. Some researchers there say that commercial publishers are now showing more interest in the region, potentially threatening its unique publishing model. And because of the high cost of APCs, some fear that Latin American researchers will not be able to afford publication in for-profit open-access journals.

Arianna Becerril-García, a computer engineer at UAEM in Mexico and the co-founder of Redalyc and AmeliCA, two of the largest open-access journal networks in Latin America, said she appreciates Plan S’ work but worries that its model will price out researchers in poorer countries. “We cannot afford, in the past, subscriptions. And now we cannot afford APCs,” Becerril-García said.

“If we forget about academia taking back control of publishing, we are going to end up in 10 years with another kind of exclusion,” she said. “We are just transferring this problem into another problem.”

For many open-access advocates, the goal is not simply to open up research to a broader readership. It is to remove scholarly inquiry from the logic of the marketplace — specifically, from a system in which publishers compete to have the highest-prestige journals, and in which academics, in turn, compete to publish in those same prestigious titles, which can be critical for career advancement. Many advocates of publishing reform complain that this amounts to what Becerril-García calls a “prestige industry.”

Björn Brembs, a neurobiologist at the University of Regensburg in Germany and a vocal critic of the publishing status quo, said that many of his colleagues are overly fixated on publishing in prestigious journals. “They could cure cancer,” he said, and still be disappointed if their work wasn’t published in the right journal. But, in his research, Brembs has marshaled data to argue that science published in the most prestigious journals is not necessarily any more reliable — and, in fact, may be less likely to replicate.

Michael Eisen, the Berkeley biologist and open-access advocate, is a fierce critic of Nature, Science, and other major journals that, he said, sometimes seem “interested in flash and sexiness rather than scientific rigor and quality.”

Eisen compares the current journal model to the Good Housekeeping seal of approval: The journal editors and peer reviewers look at a paper, and then they either accept it or reject it. Accepting it puts that journal’s brand on the paper, as a kind of guarantor of quality — a seal of approval, so to speak. Researchers are left to compete for a limited number of slots in the journals with the most prestigious seals.

“If we forget about academia taking back control of publishing, we are going to end up in 10 years with another kind of exclusion. We are just transferring this problem into another problem.”

Last year, Eisen took over as the editor-in-chief of eLife, a nonprofit, open access, scientist-led site that, according to its mission statement, aims to provide “a platform for research communication that encourages and recognizes the most responsible behaviors in science.” eLife does not have a fixed quota of papers that it accepts each year. Instead of a journal, it’s more a scholar-run platform, taking, at least in theory, any paper that meets its threshold for high quality science.

Ultimately, Eisen would like to see an academic publishing system that offers more textured assessments of new research. “We should be saying what we think about [a paper], not making a kind of hard cut,” he said. “There has to be a better, more robust and informative way for us to convey what we think about a paper, rather than just saying ‘This is or isn’t in the club.’” Instead of Good Housekeeping, Eisen imagines the future of scientific journals to be like Consumer Reports — offering detailed evaluations, available to everyone, “where we assess papers in their multidimensionality, and report back on what we think about them.”

Will that world come about anytime soon? eLife is growing quickly, but for now, it does not offer quite the textured assessment that Eisen envisions. The nonprofit also depends heavily on philanthropic funding, and it charges APCs of $2,500 per paper, waiving that fee for any lab that can’t pay.


Nearly two decades after the Budapest Declaration, a gulf remains between the messy realities of publishing and visions of reform. Globally, researchers publish approximately 2.5 million scholarly papers each year.. The work of coordinating the submission, review, and publication of that body of knowledge is enormous. Asked whether university presses could take on that role, Nick Lindsay, the director of journals and open access at MIT Press, pointed out that Elsevier likely publishes more journals than every single university press in the United States combined. He said it would require “a tremendous amount of effort to be able to capitalize the university presses to the point at which they could actually ingest all of that work.”

“The transition to digital, in the broad sense, led to this naive thinking — that everything was going to be free, everything was going to be available, everything was going to be good.”

And an open-access world poses special challenges. Indeed, part of the irony of the internet age is that, just as low-cost digital tools have raised questions about the value proposition of commercial publishers, they have also, in another sense, made those publishers seem even more essential. After all, while they may not be able to guarantee good science, major journal publishers do offer some measure of quality control. Their brands signal that journals are legitimate, and help researchers and readers sort peer-reviewed work from junk published by the rising number of predatory journals masquerading as legitimate open-access titles.

“The transition to digital, in the broad sense, led to this naive thinking — that everything was going to be free, everything was going to be available, everything was going to be good,” said Kent Anderson, a longtime publishing industry executive and consultant, and the founder of the Scholarly Kitchen, a popular industry blog. Anderson, a vocal critic of the open-access movement, sees things like the Budapest Declaration as one more piece of that early-internet-era, disrupt-the-world naivete.

Anderson rejects the idea that publishers somehow created a prestige market. Prestige, he argued, is embedded in everything from elite universities to Nobel Prizes to all the other ways researchers rank and sort themselves. It’s part of the culture of global science, and, for better or for worse, everyone in the system has to navigate it. “Guess what, people!” Anderson said. “Publishers are just part of the same market you are.”

Even if prestige does shape global science, some reformers reject the idea that the publishing status quo is somehow an inevitability. Recently, Babini has grown concerned about an open-access transition that will be dominated by the same big publishers, the same big journals, and the same system of prestige — a transition that, she fears, could displace the tradition of Latin American publishing.

It does not have to be this way, she insisted. “I am so angry!” Babini said. “It’s a lack of imagination.”


UPDATE: An earlier version of this piece described Tom Reller as vice president of communications for Elsevier. While Reller held that position at the time he provided comment to Undark, he has since left the company.

Michael Schulson is a contributing editor for Undark. His work has also been published by Aeon, NPR, Pacific Standard, Scientific American, Slate, and Wired, among other publications.