Scenarios about paying for OA

There’s an ongoing debate about ways to pay the costs of OA, especially during the transition phase between traditional business models and newer OA-oriented ones. There are several potential sources of substantial revenues. One focal point for the debate has been on revenues from academic institutions (a major source of support for knowledge dissemination), relative to revenues from funding agencies (a major source of support for knowledge generation).

In a message on the subject “Re: Failing business models“, posted to the American Scientist Open Access Forum on February 24, 2007, Heather Morrison outlined two scenarios. In Scenario A, a funding agency permits the budget of a research grant to include funds that may be used for the dissemination of research results, “let’s say $3,000“. The grantee may choose to use these funds in various ways. They may be used for the payment of the OA publication fee or APF (article processing fee) for an OA journal (or for OA articles in a “hybrid” journal, in which only some of the articles are OA) . Or, they may be used in other ways, such as to cover at least some of the costs of attending a conference to present a paper on the outputs of the research. She predicts that Scenario A would result in “competition in the scholarly publishing industry, and ultimately better quality services at lower prices“.

In Scenario B, grantees don’t need to make such choices. Instead, the funding agency permits the $3000 to be used to pay APFs for OA, but these funds cannot be used for other purposes. She notes that “For many publishers, this amount is higher than true costs of publishing an online open access article, and allows for double-dipping (revenue from both subscriptions and processing fee charges)“. She predicts that Scenario B may lead to increasing inefficiencies in the scholarly publishing industry, and an upward spiral in APFs.

In a letter entitled “CIHR’s proposal to mandate self-archiving“, published in the January 2007 issue of University Affairs, Stevan Harnad commented that “Unlike the Wellcome Trust’s self-archiving mandate in the United Kingdom, CIHR’s does not offer to fund publishing in an Open Access journal. Apparently CIHR did not feel it had the spare cash for this. This is quite understandable: all potential publication funds are currently tied up in institutional journal subscriptions, worldwide. If and when self-archiving should ever lead to institutional cancellations that make the subscription model unsustainable, then those same institutional windfall savings will be the natural source for the cash to pay for Open Access publishing. No need to take it from research funds at this time, when it is not yet either necessary or affordable“.

Note that the draft policy of the CIHR (Canadian Institutes of Health Research) can be regarded as an example of Scenario A, while the OA policy of the Wellcome Trust can be regarded as an example of Scenario B.

What might be the implications for academic institutions if they (and funding agencies) capped APFs to achieve OA at $3000 (US)? The most enlightening discussion of this issue that I’ve seen has been in a series of messages posted to the LIBLICENSE-L Mailing List on the subject “Re: Institutional Journal Costs in an Open Access Environment“. I found one of these messages, posted by Ray English on April 26, 2006, to be of particular interest. His comments were about a study that looked at “journal costs in four scientific fields for a representative group of academic and research libraries of different sizes“. The particular study that he reviewed didn’t take into account the question of grant funds supporting OA publication fees. Nor did it take into account other possible sources of revenue for open access other than publication fees. And (if I understood correctly), it was assumed that all journals would charge an APF to achieve OA, and that the only way to obtain OA to articles was via OA (or hybrid) journals (that is, OA to self-archived articles was tacitly assumed to make a negligible contribution).

His entire message should be read, but the conclusion that I reached on the basis of his comments was this one. In a scenario where the APFs to achieve OA are paid entirely from institutional (not granting agency) funds, and the fees are capped at about $3000 (US) for all journals in the four scientific fields, then “there would be substantial savings for all of the institutions in the study” in comparison with the current situation.

Furthermore, “Savings would be greatest for the comprehensive universities (around 90% or more of current costs), somewhat smaller (but still very large) for the liberal arts colleges, smaller still for the doctoral universities … and smallest for the largest research institutions. But there would still be significant savings even for the latter group“.

If this interpretation is correct, then Scenario A (outlined above) should be attractive from the perspective of institutions that currently pay substantial amounts to support knowledge dissemination. This same scenario should also be attractive from the perspective of funding agencies that have decided to support not only knowledge generation, but also (at least to a limited extent) OA-based knowledge dissemination. Of course, from the perspective of publishers, Scenario A seems likely to yield less revenue over the longer term than will Scenario B. This is because, as predicted by Heather Morrison, Scenario B might lead to less competition, and perhaps even to an upward spiral in APFs.

More evidence-based analyses of a variety of scenarios are needed.



  1. heathermorrison said

    Very good points, Jim – the study by William Walters does indeed show that almost all research-producing institutions would save money with an average article processing fee cost of $3,000 or less.
    Institutional Journal Costs in an Open Access Environment
    by William H. Walters

    This is one illustration that there is more than enough money currently going into scholarly publishing from the academic sector to pay for full open access. There are many more – see, for example, my blogpost, Elsevier Revenue and Open Access, in which I calculate that about 10% of Elsevier’s annual revenue (sure, much less than their library sales!) would pay for 460,000 open access articles published by the OA creme-de-la-creme publisher, Public Library of Science. (See the blogpost at

    What makes sense to me is that libraries, publishers, and others should begin the process of transitioning from subscriptions to open access. For the publishers involved in hybrid open access programs, like Springer, what I would suggest is, instead of paying a year’s subscription in advance, and then supporting APFs for open access, negotiate for one year of combined subscriptions / APFs. This would ensure ongoing revenue for the publisher to continue quality publishing, eliminate the possibility for double-dipping, and involve both publishers and libraries in-depth examination of the economics of transition. If some funders, like Wellcome, are able to help by paying the processing fees for their own grantees, this is very helpful to the overall process. No doubt Wellcome and other funders are negotiating carefully with publishers, keeping in mind that prices set might be charged to other researchers who do not have funding.

    Funding agencies, however, should not feel that they need to provide additional funds to cover APFs, since there is enough money in the system.

    I agree that more evidence-based analyses of a variety of scenarios is needed.

  2. tillje said

    Heather, thanks for your comment.

    BTW, the link to a prepublication copy of William H Walters’ article seems to be broken.

    However, I was able to reach a copy of the article via:

    The Google search that yielded the URL for the prepublication copy also revealed a citation for the published version (it’s not OA). The abstract is available via:
    Institutional Journal Costs in an Open Access Environment, Wlliam H. Walters, Journal of the American Society for Information Science and Technology 2007(1 Jan); 58(1): 108-120.

  3. tillje said

    A comment received from Peter Suber on March 5, 2007 and posted with his permission:

    Jim: Thanks for this useful survey of opinion. Note that three people have concluded that, under some scenarios, universities would pay more for OA journal publication fees than they pay now for non-OA journal subscriptions. William Walters is one of these three, in a section of his article that you don’t quote. All three of them use false or implausible assumptions. For my critique of their calculations, see

    Added March 6, 2007:
    Another relevant item has been posted by Peter Suber to Open Access News (March 5): More on costs to universities for gold OA.

  4. wdwalsh said

    “More evidence-based analyses of a variety of scenarios are needed.”

    I also agree and would like to note, if I may, that we are still soliciting submissions for the inaugural issue of Open Access Research.

  5. tillje said

    Stevan Harnad posted this comment (dated 6 March 2007) to the American Scientist Open Access Forum: The Siren Call of Speculations About Publishing Reform.

  6. heathermorrison said

    Stevan is baffled as to why people are talking about transition to open access publishing, rather than focusing on immediate self-archiving.

    From my perspective, scholarly communications is in the midst of the overall transition process from print to electronic. Changes in scholarly publishing are happening along the way, and indeed these are both necessary and desirable. Some of the changes relate to how work is done; much that used to be done manually is or could be automated; and some of the work that required specialized tools and/or expertise is now much more accessible to all.

    Business models for scholarly publishing have been in a process of change for many years. Even before electronic distribution, the subscriptions model was in crisis. It does not make sense to try to distribute electronic information in the same way we used to distribute print, so business models for electronic dissemination are still evolving. For example, it seems that some in the business are contemplating moving from subscriptions to usage-based pricing [a huge mistake, in my opinion, but that is a separate topic].

    It makes more sense to consider evolution of scholarly publishing towards an open access norm now, while the business is still in transition, than it would to work towards a stable, non-OA business model first, and then transition again to OA.

    As one concrete example of why this makes sense, consider the monies that are currently being “invested” in “scholarly publishing”, which are actually going into more sophisticated authentication mechanisms, designed to prevent people from reading scholarly articles. What a waste!

    From a political perspective, if the publishing industry were to embrace and support the “green” approach, financially supported by subscriptions, it might well be worthwhile wholeheartedly supporting the industry. However, as we have seen with recent statements by the AAUP, the Brussels Declaration, not to mention Nature’s revelation of the AAP’s employment of PR pitbull Eric Dezenhall, etc., many traditional publishers are fighting the “green” approach with everything they can think of. So there is absolutely no good reason to focus exclusively on the green approach.

    As a final note, I would like to note that I completely support open access mandates via self-archiving. These are no threat to the gold or open access publishing approach at all, as an article published OA can be deposited in an archive, too. The self-archiving mandate allows the most options to everyone.

    Heather Morrison
    Project Coordinator, BC Electronic Library Network [at Simon Fraser University Library]
    Please note that this comment reflects my opinion only, and does not reflect the opinion or policy of BC Electronic Library Network or Simon Fraser University Library.

  7. tillje said

    For some interesting comments about inefficiencies within the subscription market that will be reduced or eliminated in an open access world, see the message posted by Paul Peters (Head of Business Development, Hindawi Publishing Corporation), dated 28 March 2007, to the Liblicense-L discussion list on the subject “Re: Matt Cockerill’s comments“.


    The essence of a free market is that consumers (authors in the case of scholarly journals) are best positioned to decide which products provide them with the greatest value, based on their particular needs.

    So far there have been two proposed systems for how funders can make these decisions on behalf of their researchers: (1) by having a set amount (say $3,000) that they are willing to pay for the publication of an article (which is more or less the policy of the Wellcome Trust), or (2) by deciding in which journals their authors should be publishing, and negotiating individual deals with each of these “approved” journals (which is more or less the policy proposed by CERN’s SCOAP3).

    Unfortunately, neither the policy of the Wellcome Trust nor that of CERN’s SCOAP3 have a mechanism for increasing the competition between publishers, so one cannot expect that they will lead to greater efficiency in the publishing market.

    The best approach for research funders to take is simply to allow their grant recipients to include publication charges in their grant requests, just as they do for expenses related to attending a conference. This way, researchers will be able to choose how much of their research funds they would like to spend on publishing in a journal, funds which could otherwise be spent on conferences, graduate students, equipment, etc. …

  8. tillje said

    Some relevant comments have been posted by Robert Kiley (Wellcome Trust) on 3 April, 2007, as part of a Liblicense-L thread entitled “RE: Matt Cockerill’s comments [Wellcome Trust and OA fees thread]“. Excerpts:

    The Trust did consider adding an OA element to its existing grants, but decided that whilst we are in this interim phase – with many publishers now beginning to offer a hybrid OA option – it was easier (for our grantees) to simply say that these costs would be covered by the Trust.

    Keeping things easy for the researcher is important – if we lose the support of the researcher, then OA will not happen. Currently the Trust is putting a lot of effort into making our researchers aware of their grant obligations (namely to make their research outputs freely available), and the benefits of OA. Consequently, at this stage we felt that expecting them to make a value-for-money decision with regard to which journal they should publish in would be additional, unwelcome burden.

  9. 'Uthman said


    I realy enjoyed reading your blog, i needed some info on this subject for my new study economimy in the USA and your post helped me out a lot thank you for that

RSS feed for comments on this post · TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: