The price of publication

CAMBRIDGE, Mass. — Pier Paolo Pandolfi, a molecular biologist at Memorial Sloan-Kettering Cancer Center, is no stranger to the benefits of publishing in the “best” scientific journals: greater impact, wider readership—and career-boosting citations for his junior colleagues’ CVs. Of his 145 published papers, a number have appeared in the top-ranked journals: Science, Cell, Nature and its offshoots, and The New England Journal of Medicine.

So it was a radical departure last year when he decided to submit an important prostate cancer paper to a startup journal launched in October 2003. PLoS Biology had no track record or guarantee of survival, but it did have an editorial board of scientific heavyweights and a $9 million capital cushion. Moreover, Pandolfi supported the journal’s mission: By giving away its content for free, the journal’s creators hoped to shake up the entrenched and increasingly expensive world of scientific publication.

“We reasoned that contributing this paper could be a good initiative from a political point of view,” says Pandolfi, adding that he believes PLoS Biology “could be the new Science of 2010.”

Pandolfi isn’t alone in his desire to send a wake-up call to the $3.5 billion scientific journal industry. Other scientists, librarians, legislators, and even the general public have protested the exorbitant subscription costs of many of the top-tier journals. Disenfranchised populations, critics say, include scientists in poor countries and private citizens without easy access to libraries and who have to pay fees to access journals online.

In economic terms, librarians in many countries are fed up with years of journal prices skyrocketing much faster than the cost of living, to a level the libraries say is unsustainable. (As recently as 2001, for example, an annual print subscription to the journal Nature cost $650; today the fee is $1,280—almost twice as much.) With more journals being started all the time and library budgets static or shrinking, critics say publishing monopolies and high profits are hampering the mission of disseminating the fruits of the research enterprise so that other scientists can build on them to create new knowledge. Revenues suggest the scientific journal industry has benefited greatly from its citizenship in the publish-or-perish academic world. Now, some are asking, “Has the industry lived up to its civic duty?”

Open access

Scientists such as Pandolfi are all too familiar with peer review, a process many first encounter as graduate students. When a researcher completes a study, the findings are written up in a technical article and submitted to a journal for review by a panel of scientific peers. Much of scientists’ reputation and future career path rests on how often and where they publish the findings of their research. But many scientists like Pandolfi also want to share the findings of their work with colleagues and a public that may not have easy access to traditional journals. What’s more, many funding agencies now ask grant seekers to explain how they plan to share their findings with others. And therein lies the dilemma: Who should bear the cost of publication?

In part, the revolt over journal access stems from what advocates say is a fairness issue: “The public pays for our work; why shouldn’t everyone have easy, early access?” Dr. Harold Varmus asked participants in a New York Academy of Sciences briefing last winter. The outspoken Nobelist, a supporter of what many call the “open-access” movement, is cofounder of Public Library of Science (PLoS), a nonprofit organization of scientists aiming to create demand for open-access journals by publishing their own.

The launch of PLoS Biology last year and PLoS Medicine, due out this spring, represents the latest strategy of the open-access movement. Traditional journals cover the costs of reviewing and publishing scientific papers through annual subscriptions, which cost users (mainly research libraries) hundreds, thousands, or even tens of thousands of dollars annually. Access to articles is possible only through a subscribing library or by purchasing individual articles from a journal’s Web site for fees as high as $20 per article.

With more journals being started all the time and library budgets static or shrinking, critics say publishing monopolies and high profits are hampering the mission of disseminating the fruits of the research enterprise so that other scientists can build on them to create new knowledge.

PLoS Biology charges readers nothing at all. As soon as articles are accepted and posted on the Internet, they can be read, downloaded, copied, and redistributed at will—along with original data provided by the researchers. Instead of the readers, it is the authors who pay the bills: PLoS Biology collects $1,500 for each accepted paper to offset its costs in screening, reviewing, editing, and posting the articles. Authors may post their papers on their own Web sites, and they retain the copyright, rather than signing the rights over to the journal as is customary.

As with traditional journals, manuscripts submitted to PLoS Biology are reviewed by unpaid, unbiased scientists who check the papers for scientific soundness and ensure that the authors’ claims are justified by the data. Often the reviewers ask for more data or clarifications before accepting them for publication, also a common practice with traditional journal reviewers.

Changing the culture of science and publishing to an open-access world runs into much skepticism. Although Pandolfi’s prostate cancer paper ultimately was published by PLoS Biology in December, the scientist admits that his junior coauthors were resistant to the idea at first. “Convincing the postdocs was not easy,” he says. “They wanted to publish in a journal that is certified top-notch now, not in three years. One of the two first authors is going to be looking for a job soon, and he felt that if his first paper were in PLoS Biology it wouldn’t have the same weight on his CV as one in Cancer Cell.”

At Whitehead Institute, Director Susan Lindquist (a PLoS board member) gets similar responses. “I’ve had a horrible time getting people in my lab to submit papers there (PLoS Biology). They think their jobs depend on getting papers in particular journals, and they think it’s risky.”

Under review

The commercial journal publishing industry is dominated by a handful of European mega-companies that publish thousands of journals, creating a captive audience of libraries and researchers, say observers. The key players are Reed Elsevier and Wolters Kluwer in Amsterdam, Blackwell Publishers in England, and BertelsmannSpringer in Germany. All have been the target of hostility and activism, but the largest, Reed Elsevier, which publishes about 1,500 journals, including the elite publications from Cell Press, has been the most harshly criticized.

Last year, two faculty members at University of California, San Francisco called for a boycott of the Cell journals, and asked scientists not to review papers for them. The reason: On top of the $8 million that the UC system pays Elsevier for access to its electronic journals, the publisher was asking a surcharge of $90,000 a year for access to the six Cell journals, which the libraries said they couldn’t afford. After arduous negotiations, the UC libraries and Elsevier reached an agreement in early 2003, but the terms were not made public.

Other universities have made similar protests. Stanford University, Massachusetts Institute of Technology, Harvard University, the University of Connecticut, Duke University, Cornell University, and North Carolina State University all have passed resolutions criticizing high journal subscription fees. In their resolution, Stanford’s Faculty Senate also called upon faculty to consider submitting their research to open-access journals instead of traditional publications.

Customers complain that publishers use strong-arm tactics to maintain their profit margins, which at some companies have exceeded 30 percent. A common practice is “bundling” several journals that must be purchased together. An extreme case is Brain Research, which costs about $21,000 a year in a package with five other journals—you have to buy them all to gain access to just one.

The high profitability of the commercial publishers seems outrageous to many librarians and scientists, especially given the nature of the business. That outrage boiled over earlier this year when the editorial board of a computer science journal, The Journal of Algorithms, resigned en masse, charging that Elsevier makes the publication too expensive for many students and libraries. Before Elsevier took the journal over in 2001, a subscription cost $600; by 2003 it had jumped to $700, according to a statement by the editors.

Judith Messerle, a librarian at Harvard University’s Countway medical library, says that when she arrived there 15 years ago, a budget of $500,000 paid for 5,000 journal titles. Today, the library’s budget of $1.7 million buys just 2,700 titles.

Moreover, as publishers “migrate” customers to online versions of journals with their ease of access and searchability, Messerle says libraries like Harvard need to maintain their shelves of print journals as well “because the publishers don’t guarantee access to the content in the future.” So, Messerle says, if a library decided to cancel an online-only subscription, “The day you do that is the day you lose access to the content.” But to get both electronic and print journals now, she adds, the library has to pay publishers surcharges of 15 percent to 25 percent.

David Richardson, who runs Whitehead’s library, says the Institute has discontinued two-thirds of its print journals in part because of financial reasons, and also because the journals are available at the MIT library, where Whitehead scientists are on faculty. “Every five to seven years, they double in price, in my experience,” he says.

Doubts abound

The growing backlash against large publishers has stung them. But executives of these companies say they provide an excellence that justifies the high prices. And while some mainstream, subscription-based journals acknowledge the pressure for change, most are not convinced that open-access publications will survive.

“I think the numbers just don’t add up,” says Emilie Marcus, editor of the journal Cell and executive editor of Cell Press, referring to the author-pays system of PLoS. “That’s why they have a large private grant”—the $9 million startup fund from the Gordon and Betty Moore Foundation that protects them from real-market dynamics for some time. Marcus says she welcomes attempts to create an open-access industry, but believes proponents have glossed over the potential downsides of their model—such as whether the economics of the author-pays format will enable the journals to survive and maintain high standards.

Doubts also abound in the editorial offices of The New England Journal of Medicine and other leading journals about the author-pays PLoS business model. “For us, the most important thing we do is ensure the quality of the material we are publishing,” says NEJM Executive Editor Gregory Curfman. The necessary infrastructure includes editors, technical experts, manuscript reviewers, proofreaders, printers, and Web specialists. “We can’t do that at $1,500 per article,” he says, referring to the PLoS author charge. “It would be a different kind of journal.” Quality could suffer if author-pay journals have a financial incentive to accept more articles to recoup their costs, Curfman adds, potentially letting standards slide.

PLoS says it has taken great pains to hire distinguished editors (in fact, a top editor came from Cell) and to refer manuscripts to highly credible reviewers. Says Peter Suber, a professor at Earlham College and open-access advocate, “Open-access removes the barrier of price, not the filter of quality control.”

There are those who would challenge the assertion that an open-access journal industry would be cheaper for subscribing libraries than the current publishing system. At Yale University, associate librarian Ann Okerson did a quick run of the numbers, based on Yale researchers’ annual output of at least 4,400 articles. If the university paid, for example, $1,125 for each published paper, Okerson says it would cost Yale approximately $4,950,000—in excess of the estimated $3.6 million to $4 million it pays now for its science, technology, and medicine journals. Says Okerson, “It is much too early to estimate accurately the financial impact of open access.”

Removing barriers

While the debate over open access continues, national legislators, the National Institutes of Health, and private funders such as the Howard Hughes Medical Institute and the Wellcome Trust have taken steps aimed at lowering cost barriers to dissemination of research findings. HHMI has announced that it will pay the authors’ fees for its 350 investigators to publish in open-access journals such as PLoS Biology. And NIH has decreed that researchers applying for grants of $500,000 or more must describe how they will share their results—or explain why this can’t be done. In another open-access initiative, the National Library of Medicine’s PubMed Central offers free access to papers made available by willing publishers through PubMed links.

Not all journal publishers are commercial enterprises. Nonprofit scientific societies put out a large number of journals—the American Association for the Advancement of Science’s widely read publication Science, for example. The societies depend on journal revenues to support educational activities and meetings, and complain that open-access models might leave them short of revenue to underwrite these. In response, Varmus suggests that they raise these funds through means other than subscriptions.

However, according to AAAS Chief Executive Officer Alan Leshner, Science provides so much added value—such as its reporting on science news and policy, and commentaries on the papers it publishes—that it would have to charge $10,000 per paper if it adopted a PLoS-style author-pays model. As for open access in general, “We applaud the experiment, and we’re just waiting to see how this plays out.”

Some publishers, while hewing to their traditional business models, have made changes to lower barriers to access. In several cases, readers can access journal issues online free six months or a year after publication. Many, like The New England Journal of Medicine, have waived subscription fees for scientists in Africa and other countries with limited capital. NEJM research articles also are free online after six months, says executive editor Curfman. The journal, he adds, is sensitive to complaints about per-article fees from people who want access to up-to-the-minute research on diseases for themselves or their families. Curfman says, “To be perfectly honest, if someone called us up and said, ‘My daughter has disease X and needs an article,’ we’d just send it to them.” And, he adds, for individuals in or out of medicine who want the latest research news immediately, an online subscription is only $99 a year.

Still, open-access supporters such as Varmus, the PLoS cofounder who also is president and chief executive officer of Memorial Sloan-Kettering and former director of NIH, believe that over time, open-access journals will capture many of the best research papers. At the New York Academy of Sciences meeting last year he predicted that traditional publishers will be forced to change, or risk going out of business.

“If they fold, that is fine,” he said. “If they adapt by becoming open access, that is better.”