Saturday, February 24, 2007

Elsevier Revenue to Open Access

Is there enough revenue in the current scholarly publishing system to fund an open access system? This post explores what would happen if Reed Elsevier's revenue were redeployed to support for open access, and the answer is a resounding yes! For example, Reed Elsevier's 2005 revenue was sufficient to pay for over 6 million BioMedCentral articles. To illustrate how this compares with Reed Elsevier's current production, 6 million articles divided by Elsevier's approximately 2,000 titles results in an incredibly prolific over 3,000 articles per journal. Even a small fraction of Reed Elsevier's revenue could fund an absolutely amazing amount of open access. For example, 10% of Elsevier's revenue would pay for 460,000 articles in Public Library of Science; divided into Elsevier's current 2,000 titles, the result is a far above average 230 articles per journal (picture a quarterly journal with 58 articles per issue). 1% of Elsevier's revenue would be sufficient to provide hosting and support services for 141,538 journals at SFU Library, or about 70 times Elsevier's current journal production.

Details are below. For numbers and calculations, see the Elseviertooa Open Data edition:

Let's explore what happens if Reed Elsevier's revenues were redeployed to open access.

In 2005, Reed Elsevier reported total revenues of 6 million Euros, the equivalent of $9.2 billion U.S. dollars using Feb. 24, 2007 currency conversion.

If this revenue were redirected to open access, how much would it cover?

4.6 million Public Library of Science articles


6.2 million BioMedCentral articles


18.4 million Hindawi articles


14 million OJS journals hosted & supported by SFU Library


12.2 million OJS journals hosted & supported by Scholarly Exchange

How does this compare with Elsevier's current production of roughly 2,000 journals?

4.6 million PLoS articles divided into 2,000 journals: 2,300 articles / journal

6.2 million BMC articles divided into 2,000 journals: 3,140 articles / journal

18.4 million Hindawi articles divided into 2,000 journals: 9,200 articles / journal

Any of these would quality as extremely prolific journals. For example, some of the most prolific journals of the American Chemical Society average between 700 and 800 articles per year. Average journals publish a great deal fewer articles.

The amount of open access publishing that could be accomplished would be amazing with even a fraction of Elsevier's revenue. For example, a quarter of Elsevier's revenue would pay for 1.15 million PLoS articles. Divided amongst the 2,000 journals now produced by Elsevier, this means an average of 575 articles per journal per year, far above average.

Even with only 10% of Elsevier's revenue, it would be possible to publish 460,000 PLoS articles per year, or an average of 230 articles per journal assuming Elsevier's 2,000 journals. To illustrate how prolific this average is - this would be a monthly journal with an average of 19 articles per issue, or a quarterly journal with an average of 58 articles per issue.

1% of Elsevier's revenue would be sufficient to support and host 141,538 journals at SFU library, or 70 times Elsevier's current production of 2,000 journals.

Care to play with these numbers yourself? Check out:
Elsevier Revenue to Open Access: Open Data edition

This post was inspired by William Walsh's posting of summary of Elsevier's 2006 U.S. lobbying. Total expenditures: $2.84 million, an increase of 610% over 1998. (Thanks to Peter Suber on Open Access News.

Here is an update on my estimate of how many Open Journal Systems journals could be hosted and supported using Elsevier's U.S. lobbying budget alone:

SFU Library: $650 US equivalent per journal (from $750 Cdn for single journal):
4,369 journals (more than twice Elsevier's current production)

Scholarly Exchange: $750 US per journal: 3,786 journals (nearly twice Elsevier's current journal production)

[Disclosure: I work for SFU Library, one of the partners in the Public Knowledge Project which produces Open Journal Systems]. This post reflects my personal opinion only and does not represent the opinions or policy of the BC Electronic Library Network or the Simon Fraser University Library.

Update December 27, 2011: see Jeff Hecht's correction to these figures in a comment on Sauropod Vertebra Picture of the Week.
The Reed Elsevier profits quoted include other Reed Elsevier business figures, not just scholarly publishing operations. 

Jeff's comment in full:

The Poetic economics blog evidently mis-read Elsevier’s financial statement. Elsevier is part of Reed-Elsevier, which also includes Lexis-Nexis (not involved in scholarly publishing; it’s a database of legal and general publications), Reed Exhibitions and Reed Business Information. The 2009 adjusted operating profit for Elsevier is 693 million pounds sterling, or about $1096 million dollars. That’s still a LOT of money, but is only about 60% as much as you quote, so if you divide the profit by 1.5 million scholarly articles, you get $730. So the numbers don’t work out as neatly as you thought, that doesn’t invalidate your basic point.

(Full disclosure – I am a consultant and write for New Scientist, which is part of Reed Business Information, but am not an employee.)
Thanks, Jeff!

This post is part of the Transitioning to Open Access Series.

No comments:

Post a Comment

Thank you for your comment. Comments on IJPE are moderated.