Return to Nature Debates: e-access

Impacts of free access

Martin Richardson
Journals Publishing Director, Oxford University Press

The Public Library of Science's Open Letter demands that journals grant unrestricted free distribution rights through PubMed Central and similar online public resources to all original research reports they have published and this within six months of their initial publication date. What might be the consequences of complying with these demands? Along with other publishers, we at Oxford University Press have been investigating the effects of the free access that we already provide to the full text of 11 of our journals, with a delay of 12-24 months from publication. The texts are available within the Highwire Press collection, which groups together 247 sites containing about 250,000 free articles. We are also planning to make two of our best-known journals, Nucleic Acids Research and The EMBO Journal available via PubMedCentral, with free access after 6 and 12 months respectively. The aim is to determine what effect this has on usage. The initial results of the Highwire free access experiment indicate that here is no detectable increase in usage once free access is given (see figure 1).

In a recent public statement, the Public Library of Science said that two of the journals participating in PubMedCentral, PNAS and Molecular Biology of the Cell, have not lost subscribers as a result of making their content free. Our own experience, based on the 11 journals included in the Highwire archive, is that there has been an average decline of 3% per annum in institutional subscriptions to these journals in the three years since they were first published online. Before then, their average circulation had been growing.

If the scientific community wishes publishers to make their online journals available free of charge within six months of publication, then the revenues that will be lost as a result of further subscription cancellations and elimination of the opportunity to license back issues will need to be made up from other sources. This might force publishers to charge higher prices, reducing rather than increasing accessibility to the literature.

Online publication has greatly increased the dissemination of the journal literature, allowing easy access from the office, laboratory, home or airport. Unlike their printed counterparts, online journals are available 24 hours a day, 365 days a year. They are often fully searchable and are linked to other journals and books, and bibliographic and other databases. They thereby provide a wide range of services for the scientific communities they are designed to serve.

In common with many publishers, Oxford University Press currently has a policy of providing unlimited online access with each print subscription. Purchase of a single journal by a library enables all its users to access not only the current volume online, but also all online back issues, without further payment. Groups of libraries or even whole states or countries can obtain online-only access to all those journals to which they do not currently subscribe, at heavily discounted prices. We also give the authors who contribute to many of our journals free access to the online version of their article, for their own use and for use by anyone accessing their website.

The publication of a high-quality online journal service is an expensive business. In addition to all the fixed costs usually associated with print journal publishing (reviewing, editing and data processing and printing and distribution), there are costs associated with online publication (including software development costs, hosting and user support).

Selling the printed journals to libraries, individuals and other organizations pays for these extra costs. Subscriptions currently account for about 75% of our total sales revenue. This proportion has been decreasing over the past few years as libraries have continued to cancel subscriptions and non-subscription revenues have increased. An example of non-subscription income is that derived from licensing back issues for inclusion in services designed to integrate the journal archives of a wide range of publishers into a fully searchable and indexed collection of online journals and other publications that cover all subject areas, not just science.

My personal view is that a better objective would be for the scientific community, publishers and librarians to work more closely together to agree an equitable distribution of charges. The goal should be to facilitate the widest possible dissemination of online journals immediately on publication, in a cost-effective way. One solution might be for publishers to include an element of publication charges, so that the producers of research articles contribute something towards the costs of online publication. Another might be to introduce more flexible subscription pricing � for example, free access for developing countries and lower charges for infrequent users. Whatever the outcome of this debate, someone has to pay for the provision of online access to journals.

click here to see figure and legend

Martin Richardson is a Publishing Director with Oxford University Press with responsibility for the OUP's 180 journals (many of which are published on behalf of scientific societies and other organizations) and for science books and The Oxford English Dictionary.