Open access by the article: an idea whose time has come?
Scientific societies are often reluctant to embrace Open Access (OA) because they are concerned how this will affect the financial viability of their journals1. But there may be a risk-free, profitable means for learned societies to move towards the OA that many of their members want, without committing themselves to radical changes in their cost-recovery model. They have only to offer their authors, for a fair price, the option of buying OA for their articles -- authors can simply pay an extra fee to make their articles freely available online.
This idea is old2. What is new is that more publishers are trying it and discovering that they can profit from it, and that their authors welcome it. The Entomological Society of America (ESA), publisher of four major entomological journals, was the first to sell OA by the article. It began the service four years ago, and purchases have increased from 25% of published articles in 2000 to 62% last year. Its annual net revenues from OA sales over the same period increased from $17,000 to $53,0003.
The American Society of Limnology and Oceanography (ASLO) was the next publisher to adopt the model. It started in 2001 and by 2003, 66% of its authors were buying OA for their articles4. Within the past 10 months, three other well-established publishers have begun to sell OA by the article, with initial sales amounting to between 7% and 13% of articles published5.
An important question is how much publishers should charge authors to make their articles openly accessible. Both ESA and ASLO peg their charges to the prices of their paper reprints. Authors are familiar with paper reprints, and OA to articles is similar in function, but more efficient.
ASLO found, for example, that articles published with OA in 2003 were downloaded 2.8 times more often than online articles accessible only to subscribers, with this figure increasing to 3.4 times for older articles published in 20024. ASLO, concerned that the service's popularity risked reducing subscriptions to the entire contents of its journal, increased its price for OA this year from the cost of 100 reprints to that of 500 reprints, amounting, for example, to an increase from $126 to $350 for a 10-page article.
ESA has kept its price at 75% of the cost of 100 reprints, with the result that the price of OA for an 8-page article has increased from $90 in 2000 to $124 in 2004. These prices are low enough to encourage authors to buy, which in turn has resulted in substantial net revenues as noted above. When ESA adopted free access by the article, the plan was to increase the price when the service became so popular that it might cause a drop in paid access to the full contents of issues. Thus far, this drop has not happened.
The three publishers who have most recently initiated sales of OA by the article5 charge a flat per-article fee ranging from $995 to $2,160. But I would argue, based on the experiences of ESA and ASLO, that these fees are unnecessarily high. About a third of authors are yet unwilling to pay even modest fees for OA. Thus fees of no more than $500 per article may, for the time being, be better for both publisher profits and author satisfaction. Fees can be raised if revenues from restricted access are threatened. In the interim, scientific societies have an opportunity to offer their authors a desired service at attractive prices.
Scientific societies can benefit in multiple ways by offering OA by the article:
Scientific societies that offer OA by the article can increase the value of their service by maximizing the convenience with which their OA articles can be found. For example, they should make sure that the indexing robots of search services such as Google and Yahoo have access to the files of their OA articles, and they should deposit all OA articles in an Open Archives Initiative-compliant archive6.
- They can substantially increase their publication revenues without loss of income from restricted access.
- Their members are more likely to remain loyal because their societies' journals are offering a new mode of publication that many authors want.
- Their members should be pleased that their societies are perceived as helping rather than resisting the move towards a better system of communicating research results.
- Their journals should enjoy increased submissions because competing commercial journals so far do not offer an OA option.
- Their journals should compete well with new journals that are fully OA, because the former have well-established reputations and can keep their OA charges lower provided they maintain sufficient revenues from restricted access.
- As OA gains momentum their journals will be well positioned to change to new revenue models.
Many scientific societies have helped the move toward OA by making all articles in their journals freely accessible after an embargo period of six months to several years. If the embargo period is brief, the societies might assume that they will not profit from offering OA by the article. Such an assumption is unjustified. Authors and their sponsors have for example already paid $19,500 and $17,600 for OA to articles in the journals Physiological Genomics and Development, although both allow OA to all articles after one year and six months, respectively7.
The lack of rapid adoption of "self-archiving", where authors themselves make the contents of their refereed, published articles freely web accessible, suggests that most authors would prefer that publishers provide the OA and that the article be to a publisher-formatted, publisher-authenticated version8. It is time for publishers to offer authors an OA option and to benefit from it.
Thomas J. Walker9
Professor of Entomology, Department of Entomology & Nematology, University of Florida, Gainesville, FL 32611-0620, USA
- See http://www.nature.com/nature/focus/accessdebate/8.html
- See BioScience 45, 171 (1996); American Scientist 86, 463-471 (1998); Nature 411, 521-522 (2001).
- ESA publishes the Annals of the Entomological Society of America, Journal of Economic Entomology, Environmental Entomology and Journal of Medical Entomology. I have compiled statistics and a history for ESA's OA service. Estimates of net revenues from OA assume that the incremental cost of offering the service is $1 per page.
- ASLO made selected articles in Limnology and Oceanography OA starting in March 1999 but did not start sales of OA by the article until after it learned of ESA's service.
- The three publishers are American Physiological Society (Physiological Genomics), Company of Biologists (Development, Journal of Cell Science, Journal of Experimental Biology), and Hindawi Publishing Corporation (EURASIP Journal of Applied Signal Processing).
- See the Open Archives Initiative and Chuck Hamaker's posting to the American Scientist Open Access Forum.
- Physiological Genomics, published by the American Physiological Society, is the only journal offering OA by the article that does not clearly mark OA articles in online tables of contents. Beverly Ventura, Managing Editor, wrote (18 Mar 2004) that Physiological Genomics had sold access to 13 of 101 articles since the option was first offered, for $1500, in July 2003. Since January 2004, when Company of Biologists started selling OA by the article, Development sold access to 22 of 123 articles. During most of that period the service was offered for an introductory price of £500/$800. It is now £1350/$2160.
- See http://www.nature.com/nature/debates/e-access/Articles/harnad.html, http://www.eprints.org/self-faq and http://www.nature.com/nature/focus/accessdebate/4.html.
- My interest in OA to the journal literature was initiated and has been sustained by the Executive Committee of the Florida Entomological Society, which embraced OA in 1993 and has provided it to all authors in its journal since 1994.