Wednesday, 23 July 2014

Journal Prices: How Science and Science Funding Lose Out

A few weeks ago a paper in Proceedings of the National Academy of Sciences of the USA* hit the headlines. It compared the cost of journal ‘bundles’ sold by ‘for-profit’ and ‘non-profit’ publishers to university libraries in the USA, under confidential terms revealed only by freedom of information requests to state-funded institutions. Not surprisingly the commercial publishers charged more for their bundles of journals. However, the size of the bundles from different publishers obviously varied. In order to get an idea of value for money, the authors then used the citation rating of a journal to obtain the price per citation for each bundle. Not surprisingly, the cost/citation was higher for journals bought from commercial publishers than from ‘nonprofits’. They must be higher since that is why commercial publishers are in business; they are not charities.

However, in using citation indices at all, the authors sadly scored an own-goal. Citation, as a measure of scientific worth, is utterly corrosive to scientific worth. I will not repeat the old arguments here since David Colquhoun’s Improbable Science Blog continues to do a fine demolition job.

In using citation indices at all, the authors of the PNAS paper provide ammunition to the commercial publishers. They can continue to extract money from publicly-funded science while researchers feel compelled by peer or institutional personnel policies to submit papers to journals with a high impact factor. Some commercial publishers have become highly skilled at recruiting editors (usually working at least in part in their employer’s time) and editorial boards who work in popular and active, i.e. inherently highly citable, fields. A successful journal in citation terms can therefore be a moneyspinner.

The ‘nonprofit’ publisher can also be a misnomer. University presses and society journals may also be run for profit, in fact ones I have been associated with were and still are. Journal profits are used to fund other society activities like travel grants and scholarships. In Britain, journals are often set up as subsidiary trading companies covenanting profits to their parent society, thereby preserving the charitable status of the society and avoiding corporation tax on their profits. The fact that society journals do not sell for as much as commercially-published journals can probably be attributed to a lack of marketing expertise (i.e. brass neck) and general commercial nous than to a desire to keep prices low. Open publishing is as welcome to those societies that rely on journal profits to sustain their activities as it is to commercial publishers. An irony of the article in PNAS is that to see it you have to pay for it!

Until I retired I was closely involved in deciding what money should be spent on library subscriptions. The first rule in dealing with publishers is never let a librarian have any role other than in providing data. Librarians just hate to discontinue the run of a journal. The second rule is monitor usage of current journals. The third rule is to threaten to cancel a whole bundle and do so if the price asked does not fall; the price asked soon falls. The calculation we used in deciding whether to cancel other than the obviously key journals was: would it be cheaper to get the required copies of articles from the British Library than buy the journal. In this way, savings on journals could be, and was, spent employing and equipping scientists.

The PNAS article draws attention to only part of the story. Allowing publishers to charge exorbitant prices to download old articles, sometimes more than 50 years old, is an absolute rip-off. ‘Nonprofit’ scientific societies that allow commercial publishers to handle their journal sales are as guilty as the rest in this respect. 

So, until scientists stop chasing citations, until action is taken to stop publishers charging excessively for old articles, until scientists stop providing commercial publishers with their own and their employer’s services for nothing and until libraries stop buying their wares the publishers will continue to ‘gouge’†† science and science funders.

Fully open access, to publishing and reading of past works as well as present, must be the goal. And if that means changes to copyright law then get on with changing it. The PNAS paper at least shows ‘market failure’ that catch all trigger for government intervention. Action This Day is not even soon enough.


Bergstrom T C, Courant P N, McAfee, R P, Williams M A. 2014. Evaluating big deal journal bundles. Proceedings of the National Academy of Sciences of the USA , early edition doi:10.1073/pnas.1403006111

† Article in The Guardian:


†† employing the North American term, as in Jerry Coyne’s Blog at Why Evolution is True: