Abstract

The Internet has made possible the cost-effective dissemination of scientific journals in the form of electronic versions, usually in parallel with the printed versions. At the same time the electronic medium also makes possible totally new open access (OA) distribution models, funded by author charges, sponsorship, advertising, voluntary work, etc., where the end product is free in full text to the readers. Although more than 2,000 new OA journals have been founded in the last 15 years, the uptake of open access has been rather slow, with currently around 5% of all peer-reviewed articles published in OA journals. The slow growth can to a large extent be explained by the fact that open access has predominantly emerged via newly founded journals and startup publishers. Established journals and publishers have not had strong enough incentives to change their business models, and the commercial risks in doing so have been high. In this paper we outline and discuss two different scenarios for how scholarly publishers could change their operating model to open access. The first is based on an instantaneous change and the second on a gradual change. We propose a way to manage the gradual change by bundling traditional “big deal” licenses and author charges for opening access to individual articles.

Introduction

In scholarly publishing as in many other industries, the Internet has also opened innovative new ways of doing business. A grass-roots movement of scientists advocating the publication of scientific journals openly on the Web, which they called “open access,” started in the mid-1990s (Guédon 2001). Open access can be seen as part of a larger Web-enabled phenomenon of peer production, user-generated content, and open-source development (Benkler 2002, 2006). The open access advocates propose two partly complementary solutions to the problem of restricted access and high subscription prices (Harnad et al. 2004). In what they call the “gold” solution, the journals themselves become openly accessible; in the “green” version the journals remain restricted, but authors post versions of their manuscripts in either subject-based repositories or institutional repositories that are openly accessible to readers.

One interesting byproduct of the migration to predominantly electronic publishing is the wide range of alternatives concerning how restricted and open access journals can be funded (Cox 2002; Hedlund, Gustafsson, and Björk 2004). The paper format traditionally involves a subscription for the journal. The individual subscriber may be a researcher who may or may not be a member of a scholarly society. Institutional subscriptions typically are handled by the university or company library. In the case where only an individual article from a journal is needed, the local library has been able to provide document delivery services. When the online journals emerged, “big deals”—access to a bundle of journals from a publisher—to a large extent replaced the subscription model (Frazier 2001). Instead of paying for subscriptions, the libraries now pay license fees to gain access to a large number of online journals from a given publisher. In addition, the Internet also makes possible paying for downloads of individual articles.

In delayed open access the paying customers get immediate access, and full open access to the public is offered after an embargo period, usually of one year. In some fields of science it is essential to get rapid access to the latest results, so this does not decrease subscription income for the publisher too much.

In another hybrid model the journal is a normal subscription journal but authors may make their articles open access for a fee, usually about $1,500–3,000.

There are a wide variety of funding models for fully open access journals, from the community service model where the journal is operated by volunteers who put a lot of free labor into the publishing process, to professional publishing companies that fund their operations by author-side payments, in the form of either individual author payments or institutional membership fees.

Open access prevalence

The number of open access journals has risen dramatically in the past few years. In a study conducted in 2002, 317 peer-reviewed OA journals were identified (Gustafsson 2002). Since May 2003 the Directory of Open Access Journals (DOAJ) has listed OA journals and the number has grown from the initial 350 to the current 3,814 (as of January 13, 2009). DOAJ includes both peer-reviewed journals and journals exercising “editorial control.” A search in Ulrich’s periodicals directory, the most widely used directory for all kinds of journals, yields 3,438 OA scholarly/academic journals. If we also apply the criteria “refereed” in Ulrich’s, we get 2,119 journals, which we believe is the best current estimate of OA peer-reviewed journals. In addition to newly established OA journals, some existing journals have also converted to open access or offer an electronic version free on the Web.

Ulrich’s identifes 25,038 active, peer-reviewed, scholarly/academic journals (of which 16,923 are online). So OA journals make up 8.5% of all such journals, and 12.5% of all the online ones.

However, the full picture is that the OA share is much lower, since what matters to readers of scientific articles is getting access to a particular article, often because they are tracking a citation or have done a Google search; they want the article itself, not an entire journal. Most OA journals are new and not yet listed in Thomson Reuters ISI Web of Knowledge, so they tend to publish fewer articles per annum than the more established journals, partly because they get fewer submissions, partly because OA journals operated on a voluntary basis could not efficiently organize the work involved in publishing hundreds of articles per year. Björk, Roos, and Lauri (2008) did a study estimating the whole volume of peer-reviewed journal articles published globally in 2006 (1,350,000) and found that 4.6% of the articles were fully open access from the start.

Difficulties in moving towards open access

Changing the business model of scientific journal publishing to open access has proven to be much more difficult and time-consuming than most OA activists envisaged 5–10 years ago (Björk 2004). The major reason for this is that despite wide-spread enthusiasm among many of the different stakeholders in the process, this is an industry with a few dominant publishers (European Commission 2006, Competition Commission 2001). In contrast to industries where customers usually decide on one alternative out of a number of competing products, customers in this case (i.e. University libraries) have a strong pressure to buy subscriptions and licenses from all the leading publishers. The profitability of the leading publishers has usually been very good, creating little pressure to change the business model. For instance, Reed Elsevier showed operating profit margins of 35–36% during 1998–2000 (UK OFT 2002).

Even though scientists are very content with OA journals when they are readers, open access is rather low on the list of criteria they employ when deciding where to submit their papers (Parks 2002). More important is the impact factor or the prestige of the journal. The vast majority of journals with high impact factors are not open access, because such journals usually were founded before open access was a feasible alternative for journals and because the incentives to change to open access are rather low once a journal has achieved a high status. The situation is aggravated by the filtering of journals carried out by ISI Web of Knowledge. ISI usually accepts only around 10% of new candidate journals for indexing. A consequence of this is a strengthening of the current status quo of publishing structures. If, on the other hand, the big publishers moved their existing journals to open access, this problem would not exist any more, since their journals already have high impact factors and prestige and would have no problems in continuing attracting enough submissions of high quality, in contrast to the startup OA journals.

A new type of cost which universities or research funders are starting to face are author charges for publishing in OA journals (or charges for opening individual articles within otherwise subscription-based journals). It has for some time been quite apparent that authors seldom can be persuaded to pay these charges from their own research grants. OA publishers using this business model have instead gone straight to universities, offering institutional membership to buy the rights for their staff to publish a certain number of OA articles per annum.

Universities usually operate within tight budgetary limits where the amounts that can be used for library subscription costs and publication costs don’t change much from one year to the next. The key issue from the university management’s view is thus the sum total of license and subscription fees plus author fees in OA journals that they pay centrally.

Publishers might be willing to go for open access if that increased their profits or at least did not decrease them. Although some of them have at times fiercely lobbied against open access (PRISM 2008), many of them are at the same time experimenting with open access, either with individual journals or otherwise with subscription journals where authors can buy open access for their own articles. Some publishers offering the latter option have pledged that, if the total number of published papers remains constant, they will decrease subscription prices to the journals in the same proportion as the OA payments increase, in line with the decrease in the number of subscription-only articles.

In contrast to the commercial publishers, society publishers are facing a slightly different dilemma. In many cases individual subscriptions to journals are bundled with membership fees or are very cheap for members. Societies fear losing members if the journals are converted to open access, because they believe many members have joined primarily because of the cheap or free subscriptions.

The key to accelerating the move towards open access would be to get the big publishers with their large portfolios of well-established journals to change to open access. In the rest of this article we discuss two possible scenarios for how this could be achieved: instantaneous change of business model and gradual change of business model.

Instantaneous-Change Business Model

The instantaneous-change business model was first proposed by the former CEO of Ingenta, Mark Rowse, in an interview in 2003 (Hane 2003). “Imagine a publisher that has already licensed content to all the library consortia in the US. The publisher could, at a stroke, say that the license will now confer rights for the academics in those institutions to submit content rather than to access content. The publisher would have successfully flipped its business model completely, to being an open access business. So I think it’s possible to see a transition from where we are now to a completely open access world without fundamentally destroying the existing scholarly publishing business.”

Peter Suber elaborated on this idea, which he called a “Rowsean flip,” in his Open Access Newsletter (Suber 2007a). So far no traditional publisher has tried a Rowsean flip with its whole production, although some publishers (such as Oxford University Press) have experimented with changing individual titles to open access. For most publishers, there is a big commercial risk involved, and very little competitive pressure to change business models.

Instead, pressure for a Rowsean flip has recently come from the subscriber side in the area of high energy physics. A number of the biggest nuclear research institutes in the world, including CERN, have come together in a consortium, Sponsoring Consortium for Open Access Publishing in Particle Physics (SCOAP³), that aims to force the major physics journals to switch to the open access model (SCOAP³ 2007). The rationale behind the initiative is that in particle physics a few huge laboratories contribute a major part of the subscription income of the leading journals in the field. If these major clients can persuade the publishers to sell them the services of these journals as open access, there would be clear savings for the institutes compared to their current subscription costs. Such a move would also facilitate access for researchers from poorer countries. The consortium is currently collecting pledges from potential additional consortium members and had in October 2008 collected 47% of the 10 million euros they believe is required to pay for the right to publish instead of paying for subscriptions (SCOAP³ 2008). Participants in the consortium would fund participation by canceling subscriptions to the targeted high-energy physics journals.

The area of high-energy physics is a natural one to try to flip due to the familiarity of open access to high-energy researchers. Even before the Internet there was a culture of systematic exchange of preprints in this community (Kling and McKim 2000) and this has been further strengthened by the growth of the highly successful subject-specific e-print repository, arXiv.

Gradual-Change Business Model

Despite what looks like a good chance for instantaneous change in high-energy physics, gradual change is a more realistic scenario for scholarly publishing in general, in particular for the bigger multidisciplinary publishers. Oxford University Press (OUP), which has been around for nearly four centuries, is one of the major publishers actively experimenting with open access. Currently it offers “Oxford Open” to 75 journals (representing roughly a third of all its titles) and two fully open access journals. The words of the CEO of OUP testify how OUP is proceeding cautiously.

Wiley-Blackwell offers Online Open, which covers about 265 of their 1,264 journals.

Springer offers Open Choice to all of its 1,470 peer-reviewed online journals.

These experiments offer article-specific open access; they are based on the assumption that authors themselves would be willing to pay the extra charge, usually about $3,000, to “free” their articles. Authors have been slow to take advantage of this opportunity. As an example, of the articles in OUP journals that fall under this option, only 7.1% were open in 2006: 2.2% in the humanities and 11.3% in the life sciences (Richardson 2006, p. 9).

New open access publishers (PLOS and BioMed Central) found authors similarly reluctant when the publishers tried to collect publication charges directly from authors; both publishers have been trying to attract funding directly from the universities in the form of institutional membership plans. BioMed Central, for instance, has had agreements with the national library consortium of Finnish Universities, FinELib. Article-specific open access is difficult to promote, since the articles are accepted for publishing and distributed to the key readership anyway, and it is hard to demonstrate the extra value for authors and their institutions.

One solution would be for the publishers to make agreements with universities and research funders. The universities and funders would pay annual fees to provide free access to articles from all authors funded or employed by the participating institutions. The big UK-based medical research funder Wellcome Trust has made agreements to this effect with certain publishers. Although such agreements may be difficult to negotiate, they have the advantage of disconnecting author fees from the researcher budgets, much as researchers usually do not pay for general library subscriptions from their specific budgets.

At a workshop organized by the UK Joint Information Service Committee (JISC) in London in 2005, the first author of this article proposed the following approach, based on bundling subscription licenses and article-specific open-access fees in such a way that during the transition period, the publisher would not risk losing income and the subscribing universities (or consortia) would not experience a total increase in the sum of license fees and author payments (Björk 2005).

A publisher with a large portfolio of several hundred journals might currently sell a license covering electronic access to all these journals to several universities. In addition, the publisher licenses content to other organizations that have few authors submitting papers (companies, for instance), but this income is relatively minor; the bulk of the publisher’s total income is from the universities. The publisher also produces paper versions of its journals, but paper journal sales are a very minor part of the publisher’s total income. This publisher may have started to offer authors an article-specific open-access option where authors can buy open access to their own articles for a fee. The publisher now modifies its license terms for universities willing to enter into long-term agreements in which the universities continue to pay the same total license fee, but part of the payment is considered to be an institutional fee covering the article-specific open-access charges for all articles from researchers at the university.

Consider the first university or consortium signing this agreement. It has until now produced some percentage of all the articles in the journals of the publisher. After this agreement the publisher could in principle lower the subscription fees for all its subscribers by that percentage since this will be the reduction in the amount of subscribed content that customers have access to. It does not, however, reduce subscription prices overall at this stage, since the subscribing organizations in any case have access to the same overall number of articles as before. Instead it uses the savings of not reducing subscription fees to cover the article-specific open-access payments at list prices of the university signing the agreement, which effectively receives a considerable subsidy from other licensees, who have not yet signed such an agreement. Authors from universities who have not signed the deal still have to pay author charges on an individual basis or via their institutions.

As more and more universities find this agreement attractive, the move to article-specific open access will accelerate rapidly. At some point, when the number of individual open-access articles starts to reach a level of 30–40%, the pressure will mount on the publisher to lower the overall level of subscription prices to those universities who do not sign up to the bundled deal.

The publisher’s savings will, in the long run, help cover the loss of subscription income from non–author-intensive organizations. Savings will be achieved in two ways. For many journals, the paper versions can be dropped altogether. And for journals where the paper version is useful (for instance, medical journals that practitioners receive personally) the costs of the paper version can be covered by other means such as sponsorship from drug companies or low-priced subscriptions (because only the cost of the printing needs to be covered) as part of association membership fees. There will also be savings in transaction costs for marketing and subscription management.

An important factor is the speed of the transition. If a snowball effect can be achieved, the number of universities signing up will rapidly increase. Since signing the agreements should precede the subsequent rise in open-choice articles published and the corresponding drop in subscription-only content by a year or two, the problem of keeping other subscribers happy should not be overly difficult.

When the publisher eventually flips to full open access, there will be a need to restructure the way of calculating the payments of the individual university to reflect the number of articles published emanating from that university, rather than the historic institutional subscription contribution (which for e-licenses would relate to readership in the form of full-text downloads). In a realistic scenario the change in the formula should probably be gradual, and also avoid big changes in the payment from year to year. This is not a minor issue, but it is one that can be resolved.

At the workshop the proposal triggered a lively discussion and later the author had the opportunity to discuss it at length with Jan Velterop, Director of Open Access at Springer and former CEO of open-access publisher BioMed Central. Velterop pointed out that one weakness in the proposal is that it is in conflict with the pledge of publishers currently offering article-specific open access to reduce the prices of normal subscriptions in proportion to the increase of open choice (or the corresponding decrease in subscription-only content). Different models for bridging these problems and reaching agreement were later discussed in the licensing negotiations between Springer and the Finnish National University library consortium, FinELib, in the autumn of 2005, but did not lead to any concrete results.

In June 2007 Springer announced that they had signed a letter of intent with the library consortium of the Dutch Universities to explore the possibility of bundling access to the Springer journals with institutional coverage of author charges (Suber 2007b). Later in 2007 Springer also announced a licence deal with the University of Göttingen (Universität Göttingen 2007).

In October 2008 Springer purchased the leading OA publisher BioMed Central (Cockerill 2008), which publishes 195 journals. At the time of writing it is not yet known whether there will be an integration of the institutional membership scheme (for paying author charges) of BioMed Central and the big deal e-licenses which Springer offers.

Conclusions

Open access was the main topic at the 2003 International Association of Scientific, Technical & Medical Publishers conference in Amsterdam.[2] One participant in the discussion predicted that while there seems to be a widespread consensus among scientists and research funders about the merits of OA, the change will never take place. The reason is that “they” (i.e. the OA proponents) are so many and scattered that they will never get their act together. There is a lot of truth in this statement. On one hand we have a dozen big commercial and society publishers that dominate the market and are content with their current profitable way of doing business. On the other hand are thousands of research funders and universities and millions of researchers. But despite the difficulties, OA proponents are achieving results, though at a slow pace.

Things are starting to change, and this change is currently being driven by the big research funding organizations in biomedicine (i.e. the National Institutes of Health in the US and Wellcome Trust in the UK) demanding open accessibility to the results of research they fund. It is also important to remember that the publishing cultures in different fields of science differed quite a lot, even prior to the Internet, and that these differences affect how rapidly open access advances (Kling and McKim 2000).

In this paper we have in particular discussed two scenarios for how established publishers with subscription-based journals could convert to open access. The instantaneous flip is probably unrealistic except for very special conditions, such as in the field of high energy physics, where the clients are starting to require open access and are backing this with funding to buy the open accessibility of the leading journals in their field.

The second scenario is one in which a major publisher starts to bundle the license to all its journals (“big deal”) with an institutional payment for the selective open access of the articles from authors at the institution in question. The crucial issue is that this bundling is done in such a way that during the transition period the publisher is assured of the same level of revenue from the institution as before and that at the same time the overall costs for subscriptions and author charges for the institution do not increase. That way the publisher should not run great commercial risks in trying out this route towards open access.

Bo-Christer Björk is professor of Information Systems Science at the HANKEN School of Economics, in Helsinki, Finland. He is the founder of the Journal of Information Technology in Construction (ITcon), one of the earliest refereed scholarly Open Access journals, and was also involved in the establishment of a subject-based based e-print repository for construction IT papers (ITC). Between 2003 and 2008 he was chairman of the Finnish national committee for Open Access (FinnOA). Besides being an active OA spokesperson he is engaged in research concerning the scientific publishing process and how it changes due to the new possibilities offered by the Internet. Email: Bo-Christer.Bjork@hanken.fi

Turid Hedlund is assistant professor in Information Systems Science at the HANKEN School of Economics in Helsinki, Finland. She has a Ph.D. from the University of Tampere in Information Science. The theme for her doctoral thesis was cross-language information retrieval. Her current main research interests are scientific communication and the scientific publishing process, especially open access publishing. In addition to research she teaches courses in E-business and Management Information Systems and is project manager for the D-Hanken project, in which an institutional repository is being developed for her university. Email: Turid.Hedlund@hanken.fi

References

Benkler, Yochai. 2006. The Wealth of Networks: How Social Production Transforms Markets and Freedom. New Haven: Yale University Press.

Benkler, Yochai. 2002. “Coase’s Penguin, or, Linux and the Nature of the Firm.” Yale Law Journal 112, no. 3 (Dec.): 369–446. Preprint, http://arxiv.org/ftp/cs/papers/0109/0109077.pdf.

Björk, Bo-Christer. 2004. “Open Access to Scientific Publications - An Analysis of the Barriers to Change?” Information Research 9, no. 2 (Jan.). http://InformationR.net/ir/9-2/paper170.html.

Björk, Bo-Christer. 2005. “Open Access—Some Reflections from a Northern Perspective.” The Scandinavian approach: Alternative economics model study.” Presentation at the JISC International Colloquium about Scholarly Communications and Publishing, London, June 21–22, 2005. http://www.jisc.ac.uk/media/documents/events/2005/06/bochrister_bjork.pdf.

Björk, Bo-Christer, Roos, Annikki, and Lauri, Mari. 2008. “Global Annual Volume of Peer Reviewed Scholarly Articles and the Share Available Via Different Open Access Options.” Paper presented at the 12th International Conference on Electronic Publishing (ELPUB), Toronto, June 25–27, 2008. http://elpub.scix.net/cgi-bin/works/Show?178_elpub2008.

Cockerill, Matt. 2008. “BioMed Central acquired by Springer Science + Business Media.” BioMed Central Blog, October 21. http://blogs.openaccesscentral.com/blogs/bmcblog/entry/biomed_central_acquired_springer_science.

Competition Commission (UK). 2001. Report Cm5184, “Reed Elsevier plc and Harcourt General, Inc: A Report on the Proposed Merger.” July. http://www.competition-commission.org.uk/rep_pub/reports/2001/fulltext/457c1.pdf.

Cox, John. 2002. “Pricing Electronic Information: A Snapshot of New Serials Pricing Models.” Serials Review 28, no. 3 (Autumn): 171–75.

European Commission. Directorate-General for Research. 2006. Study on the Economic and Technical Evolution of the Scientific Publication Markets in Europe. Brussels: European Commission Directorate-General for Research. http://ec.europa.eu/research/science-society/pdf/scientific-publication-study_en.pdf.

Frazier, Kenneth. 2001. “The Librarians’ Dilemma: Contemplating the Costs of the ‘Big Deal’.” D-Lib Magazine 7, no. 3 (March). http://dx.doi.org/10.1045/march2001-frazier.

Guédon, Jean-Claude. 2001. “In Oldenburg’s Long Shadow: Librarians, Research Scientists, Publishers, and the Control of Scientific Publishing.” In Proceedings of the 138th ARL [American Research Libraries] Membership Meeting. Toronto, May 23–25. http://www.arl.org/resources/pubs/mmproceedings/138guedon.shtml.

Gustafsson, Tomas. 2002. “Open Access: En empirisk undersökning om fritt tillgängliga vetenskapliga journaler på Internet” [Open Access: an empirical study on freely accessible scientific journals on the Internet]. Masters thesis, Swedish School of Economics and Business Administration, Helsinki, Finland. http://oacs.shh.fi/publications/GraduGustafsson.pdf

Hane, Paula. 2003. “Stable and Poised for Growth.” Information Today 20, no. 11 (November). Available on the Web: http://www.infotoday.com/it/nov03/hane2.shtml.

Harnad, Stevan, Tim Brody, François Vallières, Les Carr, Steve Hitchcock, Yves Gingras, Charles Oppenheim, Heinrich Stamerjohanns, and Eberhard R. Hilf. 2004. “The Access/Impact Problem and the Green and Gold Roads to Open Access.” Serials Review 30, no. 4: 310–14.

Hedlund, Turid, Tomas Gustafsson, and Bo-Christer Björk. 2004. “The Open Access Scientific Journal: An Empirical Study.” Learned Publishing 17, no. 3 (July): 199–209.

Kling, Rob, and Geoffrey McKim. 2000. “Not Just a Matter of Time: Field Differences and the Shaping of Electronic Media in Supporting Scientific Communication.” Journal of the American Society for Information Science 51, no. 14 (December): 1306–20.

Parks, Robert P. 2002. “The Faustian Grip of Academic Publishing.” Journal of Economic Methodology 9, no. 3: 317–35. http://dx.doi.org/10.1080/1350178022000015122.

PRISM (Partnership for Research Integrity in Science and Medicine). 2008. http://www.prismcoalition.org/.

Richardson, Martin. 2006. “Experimenting with Open Access Publishing: An Overview.” Presentation at the Oxford Open Access Workshop, June 13, 2006, Oxford University Press London. http://www.oxfordjournals.org/news/martin%20r%20presentation.pdf.

SCOAP³ (Sponsoring Consortium for Open Access Publishing in Particle Physics) Working Party. 2007. “Proposal to Establish a Sponsoring Consortium for Open Access Publishing in Particle Physics.” http://doc.cern.ch/archive/electronic/cern/preprints/open/open-2007-009.pdf.

SCOAP³. 2008. “Three More US partners join SCOAP³.” SCOAP³ News, Sept. 23. http://scoap3.org/news/news48.html.

Springer. 2008. Springer Open Choice. http://www.springer.com/open+choice?SGWID=0-40359-12-115391-0.

Suber, Peter. 2007a. “Flipping a journal to open access.” SPARC Open Access Newsletter 114 (October 2). http://www.earlham.edu/~peters/fos/newsletter/10-02-07.htm#flip.

Suber, Peter. 2007b. “OA Project from Springer and Library Consortium.” Open Access News blog, June 22. http://www.earlham.edu/~peters/fos/2007/06/oa-project-from-springer-and-library.html.

United Kingdom. OFT (Office of Fair Trading). 2002. Report, “The Market for Scientific, Technical and Medical Journals: A Statement by the OFT.” (Sept.) Publication ref. oft 396. http://oft.gov.uk/advice_and_resources/publications/reports/media/oft396.

Universität Göttingen. 2007. “Universität Göttingen verbessert Möglichkeiten der Open Access-Publikation.” Press release nr. 257/2007, October 4. http://www.uni-goettingen.de/en/sh/3240.html?cid=2758.

Notes

1. Martin Richardson, “Experimenting with Open Access Publishing: An Overview” (presentation at the Oxford Open Access Workshop, June 13, 2006, Oxford University Press, London). For an outline of the presentation see Richardson 2006.

2. The conference, which took place May 15–16, was entitled “Universal Access to STM Information: By Evolution or Revolution?”