Search   |   Back Issues   |   Author Index   |   Title Index   |   Contents



D-Lib Magazine
May/June 2007

Volume 13 Number 5/6

ISSN 1082-9873

A Challenge for the Library Acquisition Budget


Arthur Sale
Professor of Computing (Research)
University of Tasmania

Red Line

(This Opinion piece presents the opinions of the author. It does not necessarily reflect the views of D-Lib Magazine, its publisher, the Corporation for National Research Initiatives, or those organizations who are participants in the D-Lib Alliance.)

Changes in scholarly publishing

For about a century, the relationship between research libraries and the publishers of scholarly journals was stable. Authors offered their papers free to editors; the editors validated the quality of the paper through peer review processes and improved the print layout and language; the journal was published with the selected papers, initially only in paper, but lately also online.

In parallel with this publishing process, a journal's costs were recouped through revenue earned by the publisher from individual subscriptions and research libraries' subscriptions to the journal. During most of this period, costs to libraries meant delivery of a printed copy, and its cataloguing and shelving, but lately delivery has moved to online databases and automated accession (reducing library costs). A related development has been the emergence, and now dominance, of bundling: provision of a swag of journals under an inclusive price.

There are signs of change to this scholarly publishing model as the dominant or sole one. The subscription model of journal funding is no longer universal. A significant number of journals are no longer printed on paper at all. Self-archiving of papers (or their pre-publication drafts) by authors is broadening accessibility to research. Some journals are experimenting with different funding models and accessibility arrangements.

The scholarly publishing industry

Please accept the hypothesis that rapid change is in the air. Before embarking on this analysis, let us carefully examine the scholarly publishing industry, to derive some characteristics and taxonomy. You may disagree with details, but I feel this is a reasonable summary.
  1. It is estimated that there are approximately 24,000 scholarly journals of significant quality in the world.
  2. These are produced by publishers who range in size from single-journal enterprises to large companies with many titles.
  3. The quality of these journals varies, representing a wide spectrum from extremely prestigious to the mundane. This quality measure is broadly consistent whether measured by reader evaluations or by performance-based metrics such as the Journal Impact Factor.
  4. A scholarly journal may be funded by one or more of the following mechanisms:
  1. Author-side fees, levied on the author or more likely the author's institution. Page charges are author-side fees.
  2. Reader-side fees, levied on the readers or more likely the readers' institutions. Journal subscriptions are reader-side fees, as are pay-per-view charges.
  3. Public interest contributions, provided by a meta-user of the research system. This may be subsidy by a professional organization, a national grant, philanthropy, etc.
  4. Advertising.
  1. Scholarly journals may also be characterized by the accessibility of articles to readers:
  1. Toll access: access to articles provided to those readers whose institutions have paid for it – a typical subscription funding consequence, but also applicable to pay-per-view. The articles may be in paper or online.
  2. Open access: access to any article provided free to any reader with Internet services. The articles are always online.
  3. Optional open access: a mixed model being experimented with by some publishers in which the toll access model dominates, but authors can opt to have individual articles made open access by their institution paying an author-side fee. This model may be transitional. It is sometimes called Open Choice.

As the subscription funding model and toll access come under threat from the consequences of the Internet, we have seen the emergence of the Open Access Journal – a journal all of whose articles are freely available to all readers on the Internet – whose costs are met by author-side fees or public interest contributions. The Directory of Open Access Journals [4] suggests that there are at least 2,500 Open Access (OA) journals in the world (10.4% of the 24,000). Some of them are conversions from traditional journals, and others are new operations. Examples are this magazine (D-Lib Magazine), and the stable of Public Library of Science (PLoS) biology journals. OA journals, like their traditional counterparts, show quality variation over the whole gamut.

Some toll-access journals also charge author-side fees in the form of page charges, while mainly relying on reader-side subscriptions. At present open access journals use a variety of funding mechanisms – author-side fees certainly do not apply to all, as for example, this magazine.

A challenge to libraries

The name Acquisitions Budget says it all. A century of subscription payment has resulted in librarians and academics thinking that the money they are provided for scholarly research journals is to acquire materials, and they have named the budget accordingly. This is not surprising, since it reflects the dominant thinking about subscriptions in the 20th century, and may have been the origin of the funds. However, change is on the way. Indeed the English meaning of "acquisition" is already corrupted: generally journal budgets are not expended to acquire materials but mostly to obtain a license to view.

An economist looking from outside might come up with a completely different view of the journal acquisitions budget: that it exists to fund the scholarly publishing industry. Without it, the research peer reviewing process would collapse and something else would have to be invented and funded. Researchers provide their articles free to publishers, most referees provide their services for free too, but the peer-reviewing certification and publishing processes need funding nevertheless. This applies regardless of the fact that the scholarly publishing industry is only a tiny part of the industries that benefit from research.

When the industry begins to change its funding model, the expenditure should also change to reflect that shift.

Let's do a small thought experiment. Suppose all subscription journals were to disappear (as most printed encyclopedias have done), and all scholarly journals relied on an author-side fee or public interest contributions to provide open access.

  • Firstly and most obviously, accessibility restrictions will die with subscriptions.
  • Secondly, all potential readers (most importantly all researchers) would have access to all the world's literature – a big advantage that makes this a desirable scenario.
  • Thirdly, the money now expended on subscriptions to those journals your authors read could be diverted to author-side fees to those journals in which your authors publish. As a first order estimate, the total cost to sustain the same size publishing industry would be no greater, but it should be much less. Again to a first order estimate, the relative contributions by individual institutions would not be expected to change much.

The challenge is therefore to rethink how the library 'acquisitions budget' should be expended in the present situation where 10% of journals don't charge subscription fees or in the future, when possibly 100% of journals are not subscription-based. The usual predictions are for an increasing fraction of OA journals, some or many of which want author-side fees. This may be coupled with increasing use of optional-OA (open choice) provisions. While optional-OA provisions may be highly priced at present, I believe that competition and new entrants to the publishing industry should drive the price down.

However, I should note that some have predicted a sudden collapse of the toll-access scholarly journal industry, followed by an economic re-alignment based on a lower cost structure and elimination of now unwanted services.


There are three reasonable strategies for dealing with this situation. Characterize these as the conservative approach, modest activism, and predictive behavior.

Conservative approach

One possible strategy is to do nothing. This is what most institutions are now doing. However, a 10% signal cannot be construed as being a minor issue that should be ignored. The conservative approach amounts to discrimination by the institution in favor of the subscription model and its escalating prices from decreasing subscribers, and against all alternative journal funding models. It also disadvantages some researchers who want to publish in OA journals.

Modest activism

The second strategy is to recognize the industry change, but feel unable to make any local change. To invert a popular slogan: 'Think locally, act globally'. Institutions with this mindset will keep their acquisitions budget for subscriptions, but will lobby for an increasing number of self-archived articles or lobby for research organizations to provide extra funds to pay author-side fees. The approach can be broadly characterized as fence-sitting. The problem is evident, but the institution won't do anything itself yet.

Predictive behavior

In the third strategy, the institution takes a transitional response. It recognizes that author-side fees are now a significant requirement, and moves to re-align its 'acquisitions budget' to become a 'research journal budget'. A fraction of the journals budget is reserved for supporting alternative funding models, and the institution commits to monitoring and adapting its expenditure to match the change in the industry and the activity of its authors.

Predictive strategy

It is all very well to propose that an acquisitions budget start to be transformed, but how? And how much? Consider one possible strategy.

  1. Firstly, collect the metadata for all the refereed published journal output in your institution. In Australia, this is easy since universities have to collect it anyway for the Australian Government's HERDC [2]. Elsewhere it may be in your institution's Research Data Collection. In any case, this is information that the institution really ought to have at its fingertips, so collecting it and maintaining it on an annual basis is a worthwhile exercise.
  2. Now, analyze this metadata to determine the number of articles by authors from your institution that are published in open access journals, compared to toll access journals (subscription journals). Since the quality spectrum of open access and toll access journals is broadly similar, perhaps it is not necessary to do any quality weighting. Expect to find variations by discipline.
  3. A hypothetical sociologist might say that if you are publishing x% of your articles in open access journals, then x% of your available funds should be used to support them, whether by means of author-side fees or by public interest contributions. Thus, at an institutional level, the next time you have to control journal subscription expenditure, bring it down to what is the appropriate fraction of the 'acquisitions budget'. Use the rest to establish a fund to pay author-side fees or to make public interest contributions to open access journals. Establish a mechanism for claiming fees and contributions against this fund. Do this annually.

Initially the open access journals fund may be discipline skewed, but this should settle down eventually. On the basis of numbers of journals, up to 10% of publications might be expected, which is enough to warrant taking action. However, if open access journals are not well-known or prevalent in a particular field, the percentage of OA publications will be lower.

Caveats, Comments and Conclusions

Open access

Nothing in this piece should be construed as taking away from the urgent imperative for authors to self-archive their articles in institutional repositories, and the equally urgent imperative for institutions to require them to do so. Frequently called the 'Green Road' to open access, self-archiving is within reach now and can make a big impact on accessibility of articles in a short time frame (1-2 years). Self-archiving has already focused the attention of publishers on their business models, though some are still hoping that the subscription-access model can be continued into the future. It is difficult to predict the final outcomes or the timescale.

Rather, in this piece I am focusing on how institutional libraries should begin to adapt to charging changes already evident in the scholarly publishing industry, with a leisurely transition (say five years).

Electronic access

It is already clear that most libraries don't want access to paper copies of journals any longer. Paper copies involve cataloguing and shelving, and are hardly ever used compared to electronic access to articles [3]. Some librarians have confessed to me that they are required by their contract with a publisher to purchase a paper copy of a journal in order to get electronic access to it, but that they then throw the paper copy away as the cheapest way to dispose of it. In 2007, many librarians agree that scholarly journals are almost always electronically accessed by researchers (see also section 6.4 of [3]).

The issue therefore has little or nothing to do with paper and paginated journals, except for the cost savings in not producing and disseminating paper. Time will deal with remnants of the Gutenberg era. Rather, it is all about change in the business models.


Libraries that do not adapt to the changing scenarios run the risk of being labeled by their researchers as biased and failing to meet researchers' needs. Author-side fees do not mean fees paid by authors. Researchers are entitled to expect that their institutions will fund their publications, since they are required to produce them.

Of course, it is possible to take a game theorist approach to the issue, and argue that (a) the institution gets all OA journal output for free to read, so its optimal strategy is (b) to expend all its journal funds on subscriptions to maximize available articles, and (c) encourage its authors to only publish in journals without author side fees (not synonymous with toll-access journals). Universities don't do this and it would be a very cynical approach.

Information benefit

The exercise proposed here will provide the institution the capability of knowing where its authors publish, and in what kinds of journals. It is hard to imagine that institutions don't want this information now, anyway.

How do the journals in which your researchers publish match up against those to which your institution subscribes? Is the first a subset of the second? If not, what are the differences, and can they be explained? Don't your researchers want to read other work in the places they publish? Do subscriptions for prestigious journals ('must-have' purchases) consume the entire acquisitions budget and leave nothing over for the journals in which your authors actually publish? These are interesting questions, the answers to which will vary from institution to institution.

Funder mandates

Some providers of research grants (funders of research) have recently announced that they will pay any reader-side fees associated with Open Access journals, for example the Wellcome Trust [5]. These examples have resulted in calls for all funders to do the same.

The Australian Research Council (ARC) however has firmly rejected the notion [1], noting that it has never funded publishing costs, and stating that if it were to do so, the funds could only be taken from the grants that the ARC distributed.

The ARC's stance is a reasonable one – public research institutions are already funded for basic research infrastructure, which includes the journal acquisition budget. Singling out just OA journals and optional OA journals with reader-side fees for subsidy would be an intervention by it in the publishing industry that would require very careful justification. Inevitably questions would be asked as to why this category, and not, say, sponsored OA journals or subscription journals, should be subsidized at the funder level.

When this argument has been used, some have been unable to see the difference between funder intervention and decisions by a library. The difference lies in their normal role in the journal publishing industry. A library is making normal market decisions when it purchases, and it is the library's right to decide how it spends its funds. A funder however has usually been at arms length from the journal publishing industry, and any intervention therefore needs justification.

Libraries in the publishing loop

One issue not previously mentioned is that moving to establish an open access journal fund is a step towards putting the library into the research publishing loop, in the same way that establishing an open access mandate to deposit in a digital repository does.

Libraries have traditionally heavily supported researchers as readers, but they have not done the same for researchers as authors. It is desirable for the future of libraries, and for the future of research in their institutions, that libraries become engaged in this crucial step in the research process.


[1] Australian Research Council, Draft Funding Agreement for funding commencing in 2008, Section 7.5(i), 2007. <>.

[2] Australian Vice-Chancellors Committee, Higher Education Research Data Collection Time Series Data 1992-2004, June 2006. <>.

[3] Key Perspectives Ltd, Researchers' use of academic libraries and their services - A report commissioned by the Research Information Network and the Consortium of Academic Libraries, 2007. and <>.

[4] Lund University Libraries, Directory of Open Access Journals, 2007. <>.

[5] Wellcome Trust, Open and unrestricted access to the outputs of published research, 2006. <>.

Copyright © 2007 Arthur Sale

Top | Contents
Search | Author Index | Title Index | Back Issues
Editorial | First article
Home | E-mail the Editor


D-Lib Magazine Access Terms and Conditions