The cost of subscription publishing

The following article was submitted by Stuart Lawson.

The high cost of subscription journals has been discussed endlessly among librarians and those advocating for open access. While it is common knowledge that the prices paid by libraries are higher than most can really afford, there is still surprisingly little data in the public domain about what the exact costs are. Partly this is just down to the fact that libraries haven’t traditionally published detailed breakdowns of their acquisitions expenditure, so there is no cultural norm of doing so.

Partly it is due to the contracts that libraries sign with publishers to gain access to their journals. Some of these contracts contain non-disclosure agreements which prevent librarians from publicly disclosing the prices or pricing calculations. However, while this practice does exist, it is less widespread than is sometimes assumed. In the UK the only publisher whose contract includes a non-disclosure agreement which it claims prevents signatories from releasing some data even when subject to freedom of information (FOI) requests is Elsevier. The legal position of Elsevier’s non-disclosure clause has not been tested in court and if there are any brave librarians out there who wish to pursue that route, it could be worth getting a legal opinion about whether it can be done.

Subscription costs for all other publishers can be gained by any member of the public by sending FOI requests to UK universities, either as an individual or through the website So that’s what we’ve done. The subscription costs paid by around 100 institutions to six major publishers are now openly available on figshare. We will be sending carefully-worded FOI requests separately for Elsevier data to obtain as much as is legally possible at the moment.

Transparency in subscription data is particularly important right now because we are seeing increased transparency in the price of APCs, and if this is seen without the context of the costs of subscriptions it could be used to claim that open access is needlessly expensive (thanks to Ernesto Priego for pointing this out).

Huge thanks to Ben Meghreblian for doing most of the manual labour or sending out all the FOI requests and collating the responses. Not every UK higher education institution is included in this dataset, particularly some of those institutions which have merged in the last few years, but the majority are. The notable exception is the majority of the research-intensive Russell Group universities, which were excluded because I know that someone else sent similar requests earlier this year but have not published the results yet. Even though those Russell Group universities would tend to have much higher subscription expenditure, I think it is important to see how bearing the burden of the costs of academic publishing is not limited to the more wealthy institutions.

A few caveats about the data so far:

  • The requests asked for data in calendar years, and some institutions responded with data in academic financial years. In those cases the data has been put in the column for the latter year. For example, if a figure is given for 2012/13, it is placed in the 2013 column. The money may actually have been transferred during 2012, but it will be for subscriptions for 2013.
  • Some institutions have not included expenditure through subscription agents or other intermediaries, including big deals. Others have included these costs. This makes directly comparing institutions’ expenditure more tricky.
    We are still waiting for responses from some institutions. These figures will be added to the spreadsheet as they become available.

For further details of individual requests please follow the links given in the tabs in the spreadsheet.

UPDATE: Further data has been added to the dataset, which now includes expenditure on  Elsevier journals for some universities.

18 thoughts on “The cost of subscription publishing

  1. Excellent work. Thanks as well for the kind mention.

    I know there must be more efficient ways of doing it but is it possible to get updates here (as a comment for example, or an update on the post) when a new version of the spreadsheet has been uploaded to figshare?

    1. Hi Ernesto,

      We would be happy to update the blog post or use the comments section to alert people to new versions of the spreadsheet.

      The Informed Team.

  2. Ernesto: yes, I will let the Informed team know about updates to the data and they will add a note to the post. Adding a new ‘update’ line to the top or bottom of the post will probably work best.

    Niamh: yes I have, and I am now extending his work by sending FOI requests for more complete Elsevier subscription data.

  3. Thank you very much for doing this effort. I’m currently trying to disclose the payments of all Swiss Universities to Elsevier, Springer & Wiley. So far most Universities have refused to send this information as the contracts have non-disclosure agreements . So currently I’m filling in appeals and may eventually have to go to court.

    To all libraries, please have a look at the Cornell University, which has a clear policy about Nondisclosure Clauses in licenses. And it’s true. I asked the library to send me their figures and I got them.
    Therefore it’s really disturbing that I can’t get the numbers from my local library (Zurich), which is financed by my taxes too.

    In US there already was a court case: Elsevier vs. Washington State University in 2009, where the judge decided, that the numbers alone cannot be considered as trade secret:

    Also there’s the very interesting PNAS Paper :

    One of the finding (based on 2009 data):

    “The contracts that we have seen show remarkable institution specific price variations that cannot be explained by university characteristics such as enrollment and PhD production. Some institutions have been quite successful in bargaining for lower prices, whereas others may not have been aware that better bargains can be reached. Perhaps this variation explains publishers’ desire to keep contract terms confidential.”

    1. That’s great to hear you’re trying to get this data for Switzerland, I hope you’re successful!

  4. “libraries haven’t traditionally published detailed breakdowns of their acquisitiosn expenditure”

    Not true – annual submissions on spending are made to SCONUL, albeit not at publisher level.


    1. Publisher level was what I meant by ‘detailed breakdown’. SCONUL statistics only show higher level data such as total expenditure on electronic journals.

    1. Thanks for sharing, that’s so interesting. I wonder whether the same difficulties are present in all countries? I hope that someone in each country with Freedom of Information laws tries to get the data.

Leave a Reply

Your email address will not be published. Required fields are marked *