'Publish and be damned if I pay'

Your manuscript has finally been accepted by a top journal. The proofs are on your desk - as is a bill for $600. Natasha Loder looks at page charges

The issue of journals charging to publish research papers is a touchy one. It can inspire indignant outrage from academics who find it extraordinary that they should pay to have their work published - often in the very same journal that they referee for. And on the surface it does seem strange. If no one submitted papers to journals they would have nothing to publish and nothing to sell. Many journals manage without asking academics to cough up anything from $20 to $185 a page.

Page charges are most common in the journals of American learned societies. The wide acceptance of page charges in the US is mainly due to the fact that specific research funding is available for them. In addition, such charges are seen as supporting a not-for-profit learned society. US academics can get about $1,000 per award for page charges.

Having page funds puts US researchers at an advantage and the publication of US research effectively gets a government subsidy. Although you can plead poverty, publishers often assume that if you have a grant then you can pay page charges. One granted UK academic was asked to pay Pounds 250 to publish in American Naturalist and found that arguing was useless. "I objected - there is no specific money available for page charges, so the cost has to come out of another part of the grant. They said they would withdraw the accepted manuscript and not publish it. Effectively, this is scientific censorship."

Policies do vary between research councils, but it can be difficult to get funding for page charges. The Natural Environment Research Council leaves the issue of charges to individual committees. But one committee member explains: "When we are looking to cut back on the cost of a grant, page charges are the first thing to go." Elaine Thorpe, who coordinates research grant policy at NERC, explains that the council does not address the issue of page charges, but agrees "it is something we need to think about. If researchers have to pay these charges then we should foot the bill." NERC is considering retrospective funding of page charges, but this could be costly. The Economic and Social Research Council will pay page charges but it is not something it is keen to encourage. Applicants are advised to "avoid using journals that impose such costs". Even if the costs could be found, would this be the best use of government research funding?

Academics are keen to point out that many societies manage to publish journals cheaply to society members and libraries and provide a number of other benefits without demanding page charges. Small societies, such as the British Ecological Society, also do this while keeping subscription costs very low. And if page charges are not necessary then research councils question why they should pay them. Journal publishing is already heavily subsidised by governments that provide the funding for the research they publish. So asking academics to then provide money from their own grants is seen by many as "the ultimate kick in the teeth".

But page charges are largely a way for not-for-profit societies to recover costs and make copies of the journal cheaper for members. In journals with limited circulations, page charges can yield four times as much as institutional subscriptions, whereas a larger journal might get only a fraction of its revenue from them. But publishers using these charges to keep subscription costs down are at a disadvantage when competing for submissions, particularly at a time when research funding is limited and there is great pressure to publish. So, in the face of stiff competition from commercial journals, page-charging journals have either foundered or been forced to drop charges. A three-year pilot that eliminated page charges for the American Physical Society's journal Physical Review caused a large increase in submissions and allowed it to recapture authors from a competing, non-page charging journal. It did cause it financial problems, however.

Because library finance is uncoupled from the financing of research, page charges are difficult to sustain and the pattern of journals dropping charges in the face of competition is often repeated. Authors always choose to publish as cheaply as possible.

Electronic publishing could challenge some of the existing structures. It has certainly not reduced costs, as publishers find themselves under pressure to produce both paper and electronic versions. Simple electronic journals produced from scratch could make significant savings, enabling them to challenge the positions of established monopoly journals where it would not pay for another paper journal to compete. If this were the case, page-charging journals could find themselves in difficulties. But John Haynes, publisher of a new electronic journal for the UK's Institute of Physics and the Deutsche Physikalische Gesellschaft, does not believe there are huge savings in electronic publishing. Users want additional functionality, such as searchable archives and hot-links to abstracts and articles cited - and skilled staff, and continuous software and hardware upgrades are needed to keep a web site running. Haynes's New Journal of Physics will be entirely electronic and provided free on the Internet. Costs are to be financed by "article charges" paid by authors. Haynes concedes that obtaining this charge will be "the big challenge for this journal". In the current climate it will be hard to convince researchers that charges of about $500 are in their interest. But the facts are that subscriptions are decreasing every year, causing journals to raise prices, creating further crises for libraries. If journals such as the New Journal of Physics, which is to be launched in the autumn, can provide authors with the prestige and visibility academics require, then there would be a very strong argument for asking for these charges to be covered by research funders. Research results would be very public, easily available to the taxpayers who ultimately fund the research. The irony is that by asking authors to pay article charges we have come full circle to the original US position. But if something like page charges are to work in Europe, authors will need to demand money to pay for the publication of research, while funders will need to think afresh about the best way for the costs of publishing research to be met.