Open access, publisher embargoes, and the voluntary nature of scholarship: An analysis

Shan C. Sutton

Correspondence: Contact Claire Stewart—series editor, head of digital collections and scholarly communication service at Northwestern University—with article ideas, e-mail: E-mail:

This has been a banner year for open access, and the momentum shows no sign of letting up. College and university faculties, and the granting agencies that fund a significant amount of their research, are increasingly embracing open access to ensure research findings across disciplines are freely accessible to the public and the global research community. On the college and university level, several institutional open access policies have been passed this year by faculties at schools such as Amherst College, The College of Wooster, University of Rhode Island, and Oregon State University, as well as the entire University of California system.1

On the federal level, the White House Office of Science and Technology Planning (OSTP) issued a memorandum in February directing all agencies with research and development budgets over $100 million to develop plans for requiring public access to articles and data generated by the research they fund. This directive impacts most of the major granting agencies, including the National Science Foundation, the Department of Education, and the Department of Agriculture.

Earlier this year, the Fair Access to Science and Technology Research (FASTR) Act was introduced in Congress, which would make provisions similar to the OSTP directive a part of federal law. State legislatures in California and New York also considered bills that would require public access to state-funded research. In August, Illinois Governor Pat Quinn signed into law a bill that requires each of the state’s public universities to create an open access task force with the goal of making its research available to the public online and free of charge.

Academic libraries are key contributors to the success of the open access movement. Their influence is manifested in a variety of ways, including advocating for institutional and funder open access policies, consulting with faculty on open access publishing and copyright issues, and managing institutional repositories that provide public access to faculty and student research. Institutional repositories are a primary vehicle through which “green” open access occurs by virtue of scholars depositing the accepted manuscript versions of their published articles. This method contrasts with “gold” open access in which a journal, or selected articles within a journal in the so-called hybrid model, is freely accessible on the Web, but authors are often charged publication fees instead of the publisher selling subscriptions. These article processing charges are typically found in for-profit journals but are not as common in journals published by nonprofit entities.

Within the context of green open access, there are some troubling trends in the current dynamic among institutional repositories, institutional open access policies, and reactive publisher practices that seek to blunt their impact. Many of the open access policies passed by faculty members at colleges and universities follow the Harvard model that grants the school a nonexclusive license to faculty articles upon their creation.

This license is used in conjunction with an institutional repository to make these works freely accessible online. The version of the article deposited is typically not the final published copy but rather the accepted manuscript, a post-refereed, pre-typeset copy in PDF. This model also provides automatic waivers on request to faculty who do not want a given article to be open access.

Recognizing the prevalence of open access policies at colleges, universities, and grant funding agencies, most journal publishers have adopted guidelines to articulate the conditions under which they permit versions of the articles they publish to be placed in repositories.2 These policies often involve embargo periods during which the repository version is not publicly accessible. Standard embargo periods range between six and 24 months after initial publication. Some publishers have very restrictive embargoes while others are more liberal.

Events in recent months indicate a shift toward embargoes that are longer and more constraining. This change appears to be in reaction to more widespread institutional repository implementation and open access policy adoption. It illustrates how academic institutions and some publishers are increasingly at cross-purposes when it comes to green open access.

Two changes in publisher policies this year demonstrate different but related strategies employed to limit the impact of green open access. One strategy is based on a bias against deposit in institutional repositories versus authors’ personal Web sites. The other is based on an attempt to distinguish between voluntary manuscript deposit by authors and those that are “mandated” by institutional or funder policies.

The first strategy is employed by Springer (Springer Science+Business Media), which recently altered the deposit policy for articles in its subscription-based journals. Springer previously required a 12-month embargo for accepted manuscripts placed in funder-supported repositories, such as PubMed Central, if the funder required deposit, but there was no embargo for article manuscripts in institutional repositories. Now Springer has applied the 12-month embargo to deposit in any repository, including institutional ones. Under both the current and previous versions of this policy, authors can make their accepted manuscripts freely available on their own Web sites without embargo.3

In a June 25, 2013, interview, Springer’s Eric Merkel-Sobotta stated,

We modified our policy because, with open access (both green and gold) becoming a more and more substantial part of scientific publishing, it was vital to ensure that our policies in this area were consistent and fully sustainable. For a publisher, sustainability certainly includes maintaining the revenue stream which supports the many activities that add value during the publication process.4

While embargo consistency may be a consideration in this shift, it is accompanied by an assertion of the need to protect subscription revenue streams from the threat of manuscripts being accessible without sufficient embargoes in institutional repositories.

This assumption regarding the negative impact of green open access on publisher revenue streams is not borne out by Springer’s 12 percent sales increase from 2011 to 2012 in an environment marked by expansion of institutional repositories and open access policies.5

A similar strategy can be seen in Emerald’s recent embargo policy change. In this case the target is institutional and funder open access policies. Emerald previously allowed authors to deposit accepted manuscripts in institutional repositories or their own Web sites without embargo. Emerald has now altered this policy so that a 24-month embargo is applied to any article deposited as the result of an open access “mandate” while authors who “voluntarily” deposit into an institutional repository do not face an embargo.6 This approach has previously been instituted by other publishers such as Elsevier, and has been referred to in the open access community as the “deposit without embargo if you wish, but not if you must” model.

From an economic perspective, it is interesting to note this change was in conjunction with the launch of new gold open access options in Emerald journals that will publish an article open access for a fee of $1,595 per article. As a result, Emerald authors who work at institutions with open access policies or receive research funding from agencies with open access requirements can either pay up front for immediate open access or endure a two-year embargo.

The launch of Emerald’s author-pays option, and the 24-month embargo for green open access, coincides precisely with the Research Councils UK’s (RCUK) position that gold open access is the preferred option for the research it funds, and the maximum allowable embargo for green open access to its funded research is 24 months.

This connection is not overly surprising given that Emerald is based in the United Kingdom, and the RCUK provides more than £3 billion in research funding annually, some of which goes toward gold open access fees. Emerald emphasizes its willingness to make exceptions to the embargo in cases where the author does not have funding available for gold open access, which would typically be the case in institutional open access policies. Yet the process of petitioning to do so creates an unnecessary barrier for green open access, and without approval, the default remains set at a 24-month delay in public access.

Emerald’s positions on green and gold access will be of particular interest to librarians as this company currently publishes 17 journals in the library science field.7

Publishers like Springer and Emerald are naturally looking to maximize profits by using embargoes to both protect subscription revenues from perceived green open access threats, and drive authors to their own gold options. Unfortunately, public access to research suffers as a result of longer and more stringent embargo periods. Actual data on subscription cancellations made in response to manuscripts being available in open access repositories is sorely lacking, and the related studies that do exist paint an incomplete picture. 8 The presumed harm to journal readership and subscription rates also relates to why publishers often accept authors posting article manuscripts to personal Web sites without embargo. One can only assume this is because they view these versions as sufficiently hidden from search engines (and readers) due to a lack of metadata and exposure to Web crawlers in comparison to content in institutional repositories.

Although it’s disappointing to see embargoes designed to dampen the effectiveness of institutional repositories, the related schemes to base embargo policies on a farcical juxtaposition of voluntary versus “mandated” depositing of manuscripts demands additional scrutiny. The vast majority of institutional open access policies have a voluntary essence, even if they are referred to as mandates. First, faculty members themselves voluntarily establish these policies, usually through a faculty senate or an equivalent body. Second, the policies are not coercive, and typically include waivers that are automatically granted upon request. Third, even with a policy in place, it is the authors who voluntarily deposit their manuscripts in repositories and allow them to be made freely accessible.

One can argue that there is an essential voluntary aspect to funder open access policies, as well. Grant-seeking authors should be well aware of open access requirements when they apply for funding, and in that sense they volunteer to take part in this endeavor. Scholars working under institutional or funder open access policies can obviously have their own personal desire to make their articles available through green open access vehicles, effectively nullifying publisher attempts to starkly contrast voluntary and mandated deposits. For all of the reasons outlined above, deposits of article manuscripts in response to most institutional or funder policies are voluntary in nature, and do not fall under publishers’ attempts to embargo them on the grounds of being mandated and involuntary.

These embargo trends, along with positions held by publishers, such as Elsevier and Wiley, that severely restrict the freedom of authors to deposit their manuscripts under open access policies, represent a step backwards for open access. Most institutions with Harvard-style open access policies do honor publisher embargoes, even though there is a strong argument that the pre-existing, nonexclusive license granted by the faculty to the institution takes precedent over any subsequent assertion of embargoes. The libraries that manage institutional repositories at schools with such policies respect the author-publisher relationship, and seek to expand public access to scholarly output without compromising their faculties’ publishing interests.

However, the underlying tension between the economic interests of journal publishers, and the knowledge dissemination interests of many authors, universities, and research funders, is clearly increasing through embargo policy shifts and related developments. Within this climate, all stakeholders should recognize the fundamental voluntary nature of the dissemination of scholarship, and the fact that authors, libraries, and publishers alike have the freedom to choose if and how they will contribute to the distribution of research findings through open models.

Longer, more restrictive embargoes undoubtedly delay public access to scholarship. Such policies may also inspire authors to reconsider where they publish, and motivate librarians to re-examine their institutional repository practices, in support of open access.

1. For information on institutions with open access policies. , see the Coalition of Open Access Policy Institutions at and the Registry of Open Access Repositories Mandatory Archiving Policies at
2. For information on publisher open access policies and embargoes. , see SHERPA /RoMEO at
3. See Springer’s self-archiving policy at
4. Richard, P. , “Open Access: Springer Tightens Rules on Self-Archiving,”. June.1. , 2013 ,
5. Reuters, “Springer Science Targets $1 billion in IPO, Still Eyes Sale,”. June.5. , 2013 ,
6. Emerald’s self-archiving policy can be found at, and Emerald’s author rights policies are available at
7. For a thorough examination of Emerald’s policy shift, see Poynder, R. , “Open Access: Emerald’s Green Starts to Fade?,”. June.17. , 2013 ,
8. For example, the Publishing and the Ecology of European Research project that investigated the various effects of large-scale manuscript deposits did not track journal cancellations. For more details, see A 2012 one-question survey of libraries by the Association of Learned, Professional, and Society Publishers on the likelihood of journal cancellations in cases where most of the content was made freely accessible after six months indicates there would be a major negative impact on subscriptions, but this conclusion is based on librarians predicting decisions that would undoubtedly involve a multitude of other variables, such as usage statistics, faculty demand, journal reputation, and price. For more details, see
Copyright © 2013 Shan C. Sutton

Article Views (Last 12 Months)

No data available

Contact ACRL for article usage statistics from 2010-April 2017.

Article Views (By Year/Month)

January: 6
February: 11
March: 11
April: 9
May: 5
June: 6
July: 5
April: 2
May: 17
June: 6
July: 3
August: 8
September: 17
October: 16
November: 11
December: 8