In August, the US government announced that it was adopting a policy requiring that all the research it funded is open access. A key element of this plan is that, once the policy takes effect, every research paper that results from this research must be open access the day of its publication. That means anyone can view the research—no journal subscription or one-time payment required.

That, obviously, could pose problems for the academic publishing business, which depends heavily on subscriptions as things are currently structured. To adapt to the inevitable future, many publishers have been adopting “article processing charges” (APCs), or fees paid by the people publishing the paper for the privilege of doing so. All of this is raising an awkward question: Who’s going to pay the APCs?

On Tuesday, the American Association for the Advancement of Science (AAAS) released a survey of researchers that suggests some are already struggling to find the cash to cover APCs, and in some cases, are taking it out of budgets that would otherwise pay for scientific work.

Paying the price

Research journals have a long history of charging fees for publication, going back to the days of what was called “page charges” in the days of print (fees for printing color images were also common). Combined with income from subscriptions and sometimes advertising, these offset the costs of the printing and the editors who arranged for peer review and typically left publishers with a healthy profit. For many journals, these charges have gone away with the growth of online journal access, but there was a history of fees for publication that influenced to development of APCs

As open-access journals were formed, they faced an obvious challenge: why would anyone pay for a subscription if the articles could be freely downloaded? As far as I’m aware, all of them turned to APCs as a solution. These needed to perform the same function as subscriptions—covering costs and leaving a profit—and so needed to be substantially higher than the fees previously charged to authors. Many journals that remain subscription-based have also adopted an option where researchers could have their papers made available via open access in return for an APC.

The challenge is how these APCs get paid. A number of foundations that support biomedical research have policies that enable them to pay the APCs on behalf of the researchers they fund. But many more researchers receive funding from government organizations like the National Institutes of Health and National Science Foundation. To find out how they were managing, the AAAS did a survey of US-based researchers, receiving over 400 responses.

Those responses revealed a variety of problems.

Where’s the money come from?

Even before the federal government’s open-access mandate kicks in, most of the researchers surveyed (over 60 percent) had already paid APCs, over a third of them having done so multiple times. But when it came to planning for the APC charges they faced, the numbers were roughly reversed, with 63 percent of the researchers saying they hadn’t budgeted anything for the fees. Given that, it’s not surprising that, when it came time to pay, only 10 percent or so found the process easy.

The vast majority (70 percent) took at least some of the money out of grants. About a third managed to get at least some support from their department, and about half that number managed to get funds from elsewhere in the university. Strikingly, 15 percent said they paid some of the APCs using their own money. (Numbers add up to greater than 100 percent because researchers either paid a single fee using multiple sources, or used different sources when paying more than one APC.)

The problem is that grants do not have a separate category of funding to cover APCs. As such, publishing will compete with other potential uses of the grant money: research. Nearly 80 percent of the researchers who responded said that the money for the APC would otherwise have gone to buying equipment or materials. About a third said that the APC took away from money that would otherwise have paid grad students or technicians. Another big sacrifice? Costs associated with attending conferences, which were cited by 60 percent of the researchers.

The total number of people responding is fairly small, and not everyone answered each question, so it’s difficult to know how widespread these problems are. But the issues themselves are completely predictable, given that most labs are run entirely off a single pile of money that has to pay off research and publications. And these problems, even if anecdotal, are taking place before open access becomes mandatory.

The obvious solution is to have agencies allocate some additional money to the researchers they fund to cover the cost of publishing. But this would simply shift the problem upstream, as the agencies would have to find that money elsewhere in the budget—which probably means funding less research unless they can get a budget increase dedicated to this issue.

In any case, the authors of the AAAS report paint the problem very clearly: “We face a growing risk that the ability to pay APCs—rather than the merits of the research—will determine what and who gets published.”

https://arstechnica.com/?p=1892918