House of Lords - The implementation of open access - Science and Technology Committee

The implementation of open access - Science and Technology Committee Contents

Chapter 3: Key Issues

Embargo periods

11. The Finch Group report recognised that, for the foreseeable future, there would be a "mixed economy" of green and gold open access. The report emphasised the need for care over embargo periods under the green model if subscription-based journals were to survive the transition to open access.[21]

12. Some witnesses disagreed with embargoes on principle.[22] The Government, however, recommended embargo periods of 12 and 24 months (the latter for the humanities and social sciences).[23] RCUK mandates (what Mr Willetts called a "nirvana" of) six and 12 month embargo periods in its open access policy.[24]

13. RCUK's policy, in its current form, does not make it clear that, during the initial "five-year implementation phase", the policy will allow 12 and 24 month embargo periods and that during that period RCUK "will be significantly more relaxed about ... requirements on embargoes", as Professor Rylance, Chair of RCUK, explained.[25] Indeed, as it currently stands, the policy simply refers to six and 12 month embargo periods.[26] There is significant confusion about this point.[27] For example, SAGE Publications Limited said that "signals from the funding bodies (such as RCUK) ... imply that not even lip service will be paid to the longer embargo periods required by these disciplines", and the Publishers Association have found that "when questioned on this divergence from Government Policy, RCUK has stated that 'ambiguity' is required".[28] Many others perceived RCUK's policy to be at odds with Government policy in this respect, and this led the Russell Group, for example, to recommend the phased implementation which RCUK have since clarified will be undertaken.[29]

14. Professor Douglas Kell, RCUK Information Champion, told us that RCUK will not require immediate compliance with the policy and characterised the five-year implementation phase as "a journey" to compliance.[30] This is not reflected in the text of the policy which states: "the policy applies to all research papers whose work was funded by RCUK being submitted for publication from 1 April 2013 until further notice".[31] The lack of clarity in RCUK policy and guidance, and the consequent confusion, especially given the imminent start date of 1 April 2013, are unacceptable.

15. Embargo periods for green open access in the circumstance where a journal offers gold open access but a researcher does not have APC funding have also caused some confusion.[32] To clarify the issue, the Publishers Association published a decision tree (see figure 1) which confirmed that, in this circumstance, the article should be published under the green model and deposited in a repository after a 12 to 24 month embargo period.[33] The decision tree has been endorsed by BIS and RCUK.[34] This important detail is not mentioned in the RCUK policy and guidance.[35]

Figure 1

The Publishers Association decision tree (endorsed by BIS and RCUK)[36]

16. We welcome RCUK's clarification of its stance on the length of embargo periods in evidence to us, and its willingness to be flexible about the implementation of open access. We recommend that RCUK revise section six (implementation and compliance) of its policy guidance notes to include reference to the "five-year implementation phase" and state explicitly that it will take an incremental approach to compliance in this period. Furthermore, the guidance must make reference to the Publishers Association decision tree in order to dispel the widespread confusion about embargo periods.

A "one size fits all" approach

17. The RCUK open access policy will have wide-reaching impact because the research councils fund research across a broad range of academic disciplines. RCUK's prescription of length of embargo periods (discussed above), licensing requirements and stated preference for gold open access have led to considerable opposition from representatives of disciplines who argue that their subjects need to be treated differently. Seven submissions to our inquiry criticised RCUK's policy for its "one size fits all" approach,[37] and many others raised similar concerns.[38]

18. RCUK will require articles published using APC funding to use the creative commons attribution (CC-BY) licence which allows the redistribution of the author's work (or derivatives of it) for commercial or non-commercial purposes, provided that the author is credited.[39] Some witnesses claimed that the CC-BY licence was not suitable for their own discipline, suggesting that it unnecessarily impinges on the rights of authors,[40] limits an author's ability to cite certain source documents,[41] and may cause publishers to increase their APC charges (rendering Finch economic modelling, which did not allow for a licence requirement, inaccurate).[42] But other witnesses, the Public Library of Science (PLOS) (an open access journal company), for example, argued that this view was based upon a misunderstanding of the licence.[43] Professor Rylance suggested that, in the case of the CC-BY licence requirements, it was simpler to begin with one rule but confirmed that RCUK was open to receiving specific evidence about cases where it would not be suitable.[44] We recommend that RCUK gather evidence about the suitability of the creative commons attribution (CC-BY) licence for different disciplines.

Box 2

Creative commons licence types[45]

Attribution (CC-BY)

Allows the redistribution of an author's work (or derivatives of it) for commercial or non-commercial purposes, provided that the author is credited.

Attribution-ShareAlike (CC-BY-SA)

Allows the redistribution of an author's work (or derivatives of it) for commercial or non-commercial purposes, provided that the author is credited and any new creations are licensed under the same terms.

Attribution-NoDerivs (CC-BY-ND)

Allows the redistribution of an author's work for commercial or non-commercial purposes, provided the work is presented unmodified and in whole, and that the author is credited.

Attribution-NonCommercial (CC-BY-NC)

Allows the redistribution of an author's work (or derivatives of it) for non-commercial purposes only, provided that the author is credited.

Attribution-NonCommercial-ShareAlike (CC-BY-NC-SA)

Allows the redistribution of an author's work (or derivatives of it) for non-commercial purposes only, provided that the author is credited and any new creations are licensed under the same terms.

Attribution-NonCommercial-NoDerivs (CC-BY-NC-ND)

Allows the redistribution of an author's work for non-commercial purposes only, provided the work is presented unmodified and in whole, and that the author is credited.

19. Some witnesses called for longer embargo periods for certain disciplines in order to protect journal income, necessary because of differences amongst different disciplines in the half life of articles (that is, the length of time an article is regularly read or quoted).[ 46] For example, 21 history journals have demanded embargo periods of 36 months.[47] It is argued that whereas, for some disciplines, embargo periods could be very short with no impact on journal income from subscriptions because of the short period of demand, for others, a period longer than six or 12 months would be required to protect income.[48] RCUK, however, told us that much of the debate was based on unsubstantiated assertion and there was, at present, insufficient evidence to support variations in arrangements.[49] RCUK is, nonetheless, committed to a review of open access policy in the fourth quarter of 2014 and it will, as part of the review, take into consideration any evidence about impact on specific disciplines.[50]

First mover risks

20. The Government want the UK to be a "first mover" on open access, but they recognise that there are major risks associated with this.[51] It is possible that the international direction of travel may be towards green open access or some other innovation, but BIS and RCUK have stated a preference for gold open access.[52] Australia, for example, has already expressed a preference for green open access, as has the National Science Foundation in the United States of America.[53] Figure 2 (below) shows current trends in the stated open access policy positions of some European research organisations. This suggests that there is not a clear trend of mandating gold open access in Europe.

Figure 2

Science Europe member organisations with open access positions[54]

Green open access is seen by many to have financial and other benefits. The UK has already invested sizeable funds in the associated repository infrastructure.[55] The potential of the green model, coupled with uncertainty about whether the international community will follow the UK's lead in preferring gold, means that RCUK must closely monitor international developments and respond promptly if it is to avoid "backing a losing horse". It has agreed to do so.[56] We commend RCUK's commitment to monitor international developments in open access—for example, whether gold is adopted by other countries—and willingness to amend its strategy accordingly. The Government must co-ordinate with other countries on open access policies.

Risk of unintended consequences

21. Some witnesses expressed concern about the possible impact of a shift to a "pay to publish" model, with limited funding, on the quality and standing of UK science.

LIMITS ON PUBLICATION

22. One set of worries centred on the administration of APC funds by universities. A number of witnesses argued that administrators will be forced to make difficult decisions about allocation of funds if demand exceeds supply, and might encourage academics to publish less or encourage them to publish in journals based on APC rates rather than journal quality. Furthermore, university administrators might have to select articles for publication with limited subject expertise and may choose to favour work by established researchers to protect university reputations, making it even harder for those at an earlier stage in their academic career to publish.[57] Mr Willetts dismissed these concerns on the grounds that universities and researchers have a shared goal of "academic endeavour", and because universities cared about their reputation for nurturing postgraduate researchers.[58] Nevertheless, recognising the level of the concern, he committed to keeping the issue under review.[59]

23. RCUK tried to allay fears that discipline-leading journals might not be compliant with its open access policy: "the evidence we have at the moment is that that is not going to be a great obstacle, despite people being anxious about it". RCUK agreed, however, that non-compliance would be highly undesirable and contrary to its ambition for "excellent and [the] best research".[60] Figure 3 summarises the compliance status of certain journals with RCUK's open access policy.

FIGURE 3

Compliance of selected journals with RCUK's open access policy[61]

Examples of journals not currently compliant with the RCUK policy are Applied and Computational Harmonic Analysis, the American Historical Review, Biomaterials, and the Quarterly Journal of Economics. The Publishers Association, which represents many major journals, was confident that the majority of journals would be compliant with the open access policy requirements by 1 April 2013.[62] RCUK agreed to assess any evidence of difficulties publishing in discipline-leading journals as a result of its open access policy in the 2014 review.[63]

JOURNALS

24. Two major concerns were raised relating to UK journals: impact on quality and affordability. Some witnesses suggested that the reputation of UK journals might be damaged due to a perception (unfounded or not) that editors would prefer gold articles in order to remain financially viable.[64] The Royal Society expressed a concern that quality of peer review might also be affected because of "pressure on high rejection journals (whose publishing costs per article are higher) to change their peer review standards in order to be able to set more competitive APCs".[65] Professor Rylance of RCUK argued, however, that journals would wish to protect their reputation for quality and would not risk lowering their standards.[66] When questioned about the possibility of a "race to the bottom", Mr Willetts said that the future of peer review rested on whether editorial functions were perceived to add value, and took the view that if these functions were valued, they would endure.[67]

25. Some witnesses feared that the shift to open access might lead to "predatory publishers" setting up journals with low or non-existent quality thresholds.[68] Professor Kell pointed to a website drawing attention to this type of journal and argued that the market would prevent this becoming an issue.[69] Professor Rylance added that journals had an interest in maintaining quality, and argued that one consequence of the changing landscape was that new journals could establish a reputation for high quality.[70]

26. Another worry was that international researchers might choose not to publish in UK journals if they could not afford, or were put off by, APC fees.[71] Mr Willetts was very clear that these authors would still be able to publish under the subscription model in the "mixed economy" of green, gold and hybrid open access envisioned for the foreseeable future.[72] This was another possible outcome that the Minister agreed RCUK should keep under review.[73]

COLLABORATIONS

27. Some witnesses identified a lack of clarity about who would pay APCs for articles written by UK research council funded researchers working with international contributors, or articles written by authors in different institutions.[74] Mr Willetts admitted that these details still needed to be worked out.[75]

LEARNED SOCIETIES

28. Learned societies make a valuable contribution to their respective disciplines by funding, for example, conferences, grants, fellowships, outreach work and policy advice.[76] Much of this work is funded, at least in part, by income from journal ownership or publication and some fear that, if the move to open access is too hasty, this work could be jeopardised.[77] Whilst the evidence we received from learned societies indicated their support, in principle, for improving access to research, they acknowledged that, given their range in size and present reliance on journal income, they may take longer to adapt to new business models.[78] Whilst we would not wish to recommend that the Government should distort the market in this area, we urge the Government to consider how they can support learned societies in this transition. We are pleased that Mr Willetts is meeting representatives from learned societies for this very purpose.[79]

29. It is vital that RCUK closely monitors implementation of the Finch Group recommendations to ensure that the move to open access does not damage the UK's international reputation for scholarship—both for outstanding research and globally respected journals. As a minimum, the RCUK review of open access must consider the following:

(1) whether different disciplines require different embargo periods, licences and primary models of publication, particularly in the light of evidence gathered about readership and citation half-lives;

(2) whether the UK, in stating a preference for gold open access, is moving in the same direction as other countries which are mandating open access (but not necessarily gold open access);

(3) whether article processing charges have adversely affected the number of international articles published in UK journals;

(4) effects on the quality of peer review;

(5) impact on the number of collaborations by UK researchers; and

(6) effects on learned societies.

RCUK must remain vigilant beyond the planned 2014 review. We recommend that it commit, as a minimum, to a further review of the implications of its open access policy in 2016 and an end-stage assessment in 2018.

Cost benefit analysis

30. The funding for APCs has been taken out of the fixed pool of funding for scientific research in the UK. RCUK argued that this is justified, given that dissemination of results is a critical part of research activities.[80] Furthermore, Mr Willetts is persuaded that open access will offer benefits to researchers, businesses (including small and medium sized enterprises) and the public.[81] The Green Book, Her Majesty's Treasury's framework for the appraisal and evaluation of all policies, states that benefits of a particular policy "should be valued unless it is clearly not practicable to do so".[82] The Finch Group undertook economic modelling about the possible costs of open access, and studies, both by BIS and external organisations, have considered possible costs of the move.[83] We were surprised, however, that the Government have not undertaken a full cost-benefit analysis of this policy,[84] and seem to have based their decision on international trends,[85] assumed benefits to the public and industry,[86] and a desire to be a world leader.[87] We recommend that the Government undertake a full cost-benefit analysis of the open access policy, particularly given the current economic climate and the consequent pressures on the public purse. This analysis must be updated to reflect actual rather than projected costs during the transition period.

Consultation

31. Effective consultation and engagement with relevant stakeholders are central to the success of the open access policy given that it is a radical change for some and affects the interests of many.[88] Unfortunately, RCUK's engagement with academics and publishers in developing its policy appears to have been unsatisfactory. The Association of Learned and Professional Society Publishers, for example, said that the policy was "presented as a 'done deal', with no room for consideration of individual subject areas, researchers' requirements and needs, or indeed the potential harm that could be imposed on the scholarly communications process".[89] The International Society for First World War Studies said that "policy-makers, including RCUK, have so far systematically refused to engage with their critics in any sustained and constructive fashion".[90] Frustration about consultation was almost universal,[91] but the Publishers Association were the most critical: "RCUK has acted unilaterally and in isolation".[92]

32. RCUK has attended various stakeholder meetings in recent months,[93] and circulated a draft policy to certain groups before publication in July 2012. It did not, however, hold a public consultation and most of the discussions about the policy have taken place since its release.[94] Professor Rylance's primary defence of RCUK's approach to consultation was that the policy was a draft document for discussion.[95] It is a matter of some concern that RCUK has since stated that it is only prepared to redraft the accompanying guidance notes, not the policy itself.[96] We note, however, that, importantly, RCUK is prepared to show more leniency over compliance than its policy implies and intends to review the success of the policy in 2014. Nevertheless, misunderstandings about the policy and perceptions of disenfranchisement are significant causes for concern which need to be addressed.

33. In the light of the significant confusion and perceptions that RCUK, at worst, "acted unilaterally", or, at least, consulted inadequately in devising its open access policy, we recommend that BIS undertake a review of how RCUK consulted over this significant change in policy with the scientific and publishing communities, to ensure that lessons are learnt.

21 Op. cit. Accessibility, sustainability, excellence: how to expand access to research publications. Back

22 Clobridge Consulting, Dr Tom Olijhoek, Public Library of Science (PLOS), Dr Daniel Turner. Back

23 Op. cit. Government Response to the Finch Group Report. Back

24 Q 71. Back

25 QQ 53-54. Back

26 Op. cit. Research Councils UK Policy on Access to Research Outputs. Back

27 Academy of Social Sciences, British Academy, British Psychological Society, the Publishers Association. Back

28 SAGE Publications Limited, the Publishers Association. Back

29 The Russell Group, QQ 53-54. Back

30 Q 66. Back

31 Op. cit. Research Councils UK Policy on Access to Research Outputs. Back

32 Institute of Physics, Society of Biology. Back