RE: Summary paper from the Publishing Research Consortium

["David Prosser" <david.prosser@bodley.ox.ac.uk>]

The Beckett and Inger paper 'Self-Archiving and Journal 
Subscriptions: Co-existence or Competition?' gives us a 
hypothesis (p. 11 of the summary paper):

'In the extreme case of 100% availability of content on the 
institutional archives and a 24-month embargo, still nearly half 
the market for subscription journals has disappeared.'

So, if 100% of the journal's content is freely available the 
journal will, all other factors being equal, lose a massive 
proportion of its subscription base.  Decreasing the embargo to 
zero increases the predicted fall in the market from 50% to 
approximately 70%.

Can we test this hypothesis?  If we look at journals hosted by 
HighWire Press we can see that a large number make papers freely 
available after 6, 12, or 24 months (see 
http://highwire.stanford.edu/lists/freeart.dtl). For these 
journals, the final versions of papers are made available to all. 
If the prediction made by Beckett and Inger was true then these 
journals should have started to haemorrhaging subscriptions 
following the opening-up of the archives.  Is there any evidence 
that they have?

Back in 2005, John Sack wrote, in a history of HighWire Press
(http://www.ingentaconnect.com/content/alpsp/lp/2005/00000018/00000002/art00008)

After several years of content was online, Nick Cozzarelli 
(PNAS), Bob Simoni (JBC) and Michael Held (Rockefeller University 
Press) presented a concept of 'free back issues' to their 
colleague HighWire publishers. Their view was that librarians and 
researchers were subscribing because they needed access to 
absolutely current issues, and that there was significant 
educational benefit in issues that were months old. They proposed 
that back issues (6 or 12 months old) be made freely available to 
the public to support educational uses, and expected that this 
would have no significant effect on subscription count. Gradually 
more and more journals came to this same belief, and today the 
programme comprises the largest archive of free full-text 
research articles that we know of: over 825,000 articles from 
about 220 journals.

There does not appear to be a mass retreat from the free back 
file programme - are publisher sanguine in the face of 50% 
declines in their subscription base?

Of course, most of the HighWire hosted journals offering free 
backfiles are in the biological and medical fields, but as the 
summary does not break-down the response of librarians by subject 
area, it is difficult to tell what predictions are being made in 
these fields.

So, we have a hypothesis and we have some test-cases. If the 
HighWire-hosted journals are managing to survive despite the 
predicted massive falls in subscriptions they should have 
experience, why should we take the Beckett and Inger study as a 
credible warning of what might happen as self-archiving become 
more widespread?

David C Prosser PhD
Director
SPARC Europe
E-mail:  david.prosser@bodley.ox.ac.uk
http://www.sparceurope.org