Business, Innovation and Skills CommitteeWritten evidence submitted by the Department for Business, Innovation and Skills

I am writing in response to your announcement on 18 January 2013 of your Committee’s Inquiry on Open Access.

I wrote on the 18 January 2013 to Lord Krebs in response to his letter concerning the House of Lords Science and Technology Committee’s inquiry. I have attached this letter as an annex as it is of direct relevance to your inquiry.

I understand that the specific issues of interest to your own Committee are as follows:

The Government’s acceptance of the recommendations of the Finch Group Report “Accessibility, sustainability, excellence: how to expand access to research publications”, including its preference for the “gold” over the “green” open access model;

Rights of use and re-use in relation to open access research publications, including the implications of Creative Commons “CC-BY” licences;

The costs of article processing charges (APCs) and the implications for research funding and for the taxpayer; and

The level of “gold” open access uptake in the rest of the world versus the UK, and the ability of UK higher education institutions to remain competitive.

I have addressed these issues here in turn.

Government’s Acceptance of the Recommendations of the Finch Group Report

Government moved quickly to respond to the Finch Report to ensure that the momentum created by the constructive Finch Group process was not lost and, as importantly, to proactively influence thinking in Europe and elsewhere around the world.

In keeping with the over arching transparency agenda, Government had a preference for easier taxpayer access to publicly funded published research as set out in BIS’ “Innovation and Research Strategy” (December 2011). Therefore, Government accepted all of the Finch Group’s recommendations, aside from the requested exemption from VAT for digital publications (which is determined by EU policy). This positive response was set out in my open letter to Dame Janet Finch (see Annex A).1

Whilst fully accepting the case for Green OA as part of the “mixed-economy” described by the Finch Group, the Government’s preference is for Gold OA, which is now being introduced by RCUK. Gold OA is considered to be an honest and explicit recognition of the cost to publishers and Learned Societies of the value they add to the translation of knowledge including through peer review and wider services to research and scientific communities.

Gold OA enables information to be made immediately accessible. Under Green OA information is denied to users during embargo periods of months or even years. CC- BY licence conditions allows for the greatest possible greatest utilisation of research.

Rights of use and Re-use Including the Implications of Creative Commons (CC-BY) Licences

Creative Commons copyright licences provide a simple, standardized way to grant copyright permissions to creative works. Of all the CC licences, the CC-BY licence provides the greatest freedom to users to search, down load and re-use the information. RCUK have expressed a firm requirement for CC-BY licence conditions to apply where there has been payment of an Article Processing Charge (APC) for publication on a Gold OA basis. The Government supports RCUK’s requirement as we believe the CC-BY license provides the greatest potential for users. It opens up the prospect of new value added services being created by publishers and others for the benefit of end users including the public, SMEs and other business users. We will improve global access to the UK’s research base and the productivity of the research process.

The Costs of Article Processing Charges (APCs)

Article Processing Charges (APCs) arise only in the case of Gold OA. They are paid up-front to the publisher or Learned Society for the services they provide. The cost of facilitating the peer review process and sustaining the infrastructure needed for publication of the final version of record by publishers or Learned Societies that opt for the Green OA approach are recovered through the subscription charges or access fees they charge during the embargo period.

Government recognises that the publication of research is a legitimate cost and forms an integral part of the research process. Nevertheless, Government also considers the research publications market too complicated and dynamic for Government to attempt to set the price of an APC. The market will decide and the market is changing.

The price of APCs will be determined by a number of factors including the quality of the service or product (from the publisher), the brand value of the journal publication, the segment of the research publications market being targeted by the publisher, the perceived value added by the publisher, the publications policy or strategy of the HEI concerned and the wishes of the customer (the researcher). There is evidence emerging in the literature that the move to Gold OA and the payment of APCs promote change in the research publications market and increase the level of competition amongst providers.2

In choosing their research publication channel, HEIs/researchers will be expected to strive to comply with RCUK’s stated OA policy objectives (such as using Gold OA or Green OA with embargo periods of no more than six–12 months).3 However RCUK and Government recognise that we are on a journey. The policy framework for this transitional process is as published by Government in response to the Finch Report and illustrated in the decision tree on the Publishers Association website.4 Researchers and HEIs will be expected to publish in journals that comply with RCUK’s policy but they will remain free to choose which publication serves their interests and requirements best. Where a researcher is unable to find a journal that complies with RCUK’s policy publication on a non-compliant basis will be possible. Publishers will need to set their APCs at a level that allows them to effectively compete.

RCUK have estimated that perhaps only 45% of the papers resulting from their funded research will be published by HEIs and their researchers on a Gold OA basis in 2013–14. Based on an average APC of £1,727 excluding VAT (as estimated in the Finch Report), and providing funding at 80% of full economic cost for a population (based on 2010/11 figures) of around 26,000 research papers published by HEIs from RCUK funded research, RCUK are allocating £17 million for block allocation to HEIs in 2013-14. This will rise to £20 million in 2014-15. Block funding will be in proportion to known labour effort expended on research projects by individual HEIs. Thirty HEIs have already been allocated a share of an initial £10 million made available through RCUK this financial year (2012–13) to pump prime preparations for setting up Publication Funds. RCUK are now sharing any best practices.

Subject to how the market develops and the outcome of the next Spending review, RCUK expect to spend a total of some £100 million on Gold OA over the first five transitional years. In addition to these funds HEFCE have stated that HEIs will be free to use part of their QR funding (which totals about £1.9 billion per annum) to supplement RCUK’s contribution to HEI Publication Funds. Hence, funders expect to be able to satisfy the level of demand for Gold OA.

There is an opportunity cost to the provision of these Publication Funds. They will use money that would otherwise be spent on research. However, independent estimates by Finch, Wellcome and BIS show that the cost of Gold OA, that is the total opportunity cost, will only be about one% of the science budget of £4.6 billion per annum.

All figures included here can only be estimates. It is not certain as to precisely how the OA market will develop both in terms of demand for and supply of Gold OA.

Level of Gold Open Access Uptake in the Rest of the World and the Global Competitiveness of UK Higher Education Institutions

The UK research base already does punch above its weight—with one percent of the global population, the UK produces about seven% of the world’s published research articles and receives some 11% of global citations. The UK’s share of the world’s top one% of most highly cited papers, an indicator of the quality of research, was 13.8% in 2010, second only to the US.5 A move to OA and particularly an OA policy with a strong preference for Gold OA should strengthen the UK’s position even further.

Publication of research articles on an OA basis increases the prospect of other researchers as well as business and public users gaining access to the published article. Citations for UK research papers will therefore increase. Gold OA ensures immediate access, further increasing the prospect of UK based research being cited and exploited. A move to OA and particularly Gold OA will enhance, not detract from, the importance of UK research around the world.

The number of OA articles in 2000 has been estimated to have been only 20,000, but over the past decade the position has been rapidly changing. BIS’ own economic analysis (at Appendix 2) showed that the OA market was already growing prior to the UK’s reforms. Government’s OA policy and RCUK’s implementation of it will provide an additional “nudge” to the process. As well as expanding access to published research this “nudge” should incentivise the UK’s research community and publishing industry, including the UK’s Learned Societies, to position themselves to exploit the global opportunities that lie ahead.

A recent study of global trends by Mikael Laakso and Bo-Christer Bjork 6 has shown that in 2011 the number of scientific articles published by 6,713 full immediate OA journals (equivalent to Gold OA) had grown to 340,000 from 288,000 in 2010. Of these, around 49% required Article Processing Charges (APCs). This research suggests that 17% of the 1.66 million articles published globally during 2011 were OA. Within this 17%, 11% were immediately available (Gold OA) and six% made publicly available within 12 months of publication (Green OA).

The UK appears to be both leading a global trend but also going with the grain of the academic community and increasingly the publishing world. Other countries are increasingly moving to OA policies. Europe accounts for around 37% of global OA articles and North America for about 19%, but Asia is also responsible for 25% of global OA, and the proportion of Gold OA is growing as stated above. As I said to Lord Krebs, the fear that the UK may be acting alone is unfounded. There are the following international developments:-

The European Research Council has operated an OA policy since 2006 for its awards.

OA will be a key feature of Horizon 2020 and will apply to all funding.

The European Commission has issued a recommendation to Member States on improving their OA policies and practices. This allows for both Gold and Green OA.7

Science Europe is supporting the development of coordinated policies across Europe with a clear aim ultimately to replace a reader-paid publication system with an author- or institution-paid one.8

Deustche Forschungsgemeinschaft (DFG)9 introduced its current OA policy in 2006. DFG expects the publications issued from research it has funded to be made available in OA within six to 12 months Costs to cover APCs are eligible as research costs within the grant.10

FWF, the Austrian research funding agency, has had an OA policy since 2004 for sciences and humanities, and commits to the payment of APCs as a research costs.11 In 2011, FWF spent around €1.5 million on APCs, for both Gold and hybrid journals.12 FWF policy states that publications should be available via OA after six months.

In the USA, NIH has recently announced new measures to enforce its OA mandate, which requires all peer reviewed research papers to be archived in PubMedCentral within 12 months at the latest.13 A bill is proposing to expand this to all Federally funded research.14

Fast growing nations such as India and China also have OA on their agenda. The focus of discussion is on shortening embargo periods to a maximum of 12 months and enabling maximum use and re-use of published research as in repositories (COAR, the repository directory, currently lists 33 repositories for China and 52 for India).

Several countries in South America have implemented legislation supporting OA.15

The newly formed Global Research Council has made OA its main agenda item for their meeting in May. The aim is to develop and agree an action plan for implementing OA16

The Australian Research Council recently announced its OA policy and cites coming in line with other international funding agencies, such as the UK, as a key objective.17

It will be evident that the UK is not alone and is managing the process of change. RCUK are now working closely with HEIs to ensure that they are able to adjust flexibly to this new OA research publishing regime. I have already arranged to hold a meeting with stakeholders on 12 February to hear about what they feel is going well and to discuss any residual concerns that they may have. There will be a progress meeting with Janet Finch’s Group by the autumn to gauge how the more detailed implementation arrangements, being led by RCUK, are progressing. In addition, RCUK have promised to hold a review in 2014 to assess how developments during the first part of the five year transitional period compare to their expectations.

I look forward to further discussing these points with you and your Committee.

The Rt Hon David Willetts MP
Minister for Universities and Science

8 February 2013

APPENDIX 1

LETTER TO LORD KREBS, CHAIR, SCIENCE AND TECHNOLOGY SELECT COMMITTEE, HOUSE OF LORDS

Dear Lord Krebs,

Open Access

I am writing in response to your letter of 24 December in which you set out your Committee’s plans to hold a short inquiry on the subject of Open Access (OA). I welcome the timely initiative that you have taken to discuss this important issue as it enters the implementation of policy phase and will be pleased to accept your invitation to give oral evidence on Tuesday 29 January. I trust that you will accept this letter as the written response that you requested by Friday 18 January.

In your letter you have asked me to address four key areas:

Support for universities in the form of funds to cover Article Processing Charges (APCs) and the response of universities and other HEIs to these efforts;

Agreeing embargo periods for articles published under the Green model;

Engagement with publishers, universities, learned societies and other stakeholders in the development of Research Council OA implementation policies; and

Challenges and concerns raised to date by stakeholders and how these have been addressed.

I understand that your Hearing on 15 January with Dame Janet Finch afforded you an excellent introduction to the subject. The Finch Group was brought together by BIS as an independent group. Janet Finch has thus provided a comprehensive explanation of the process that the Finch Group went through to arrive at its recommendations. Hence, in this written evidence BIS has focused on these four issues in turn.

Support for Universities and HEIs

The Government recognised in its response of the 16 July 2012 to the Finch Report (see Annex A) that a Government preference for a Gold OA policy, with the merit of providing free, immediate and unrestricted access and use of published research by the reader, would not come at zero cost. It was accepted that publishers, including Learned Societies, provide a valuable service by both administering a highly respected peer review process and providing the necessary investment and infrastructure for a globally successful UK based research publishing industry. To sustain the advantages of the present publishing arrangements, whilst simultaneously exposing the industry to radical change, requires payment of an Article Processing Charge (APC) under Gold OA. Open Access (OA) needs to be paid for in some way, and given the Government’s policy of a strong preference for Gold OA and permitting Green OA (which places the financial burden on users), the Government has facilitated funding for the implementation of Gold OA in two stages.

In the first stage, designed to quickly pump prime the creation of Publication Funds and to identify any “teething problems” with their formation for subsequent more long-standing and systematic arrangements, an initial one-off sum of £10 million was allocated to support “Gold OA”18 publication of research. This was announced on 7 September 2012 (see http://news.bis.gov.uk/Press-Releases/Government-invests-10-million-to-help-universities-move-to-open-access-67fac.aspx ). It is providing initial funding for the group of HEIs selected by the Research Councils, as listed at Annex B, allowing them to make preparations for the introduction of OA and to identify issues that will need to be addressed by RCUK in their implementation guidance such that the wider group of HEIs benefit from their experience.

The funding has provided some limited, but immediate, support to enable research intensive HEI’s to begin to plan for and implement Gold OA in preparation for the change that will take effect from 1 April 2013.

The initial £10 million was intended to be a pre-cursor to a more substantive, systemic and on going funding allocation to be made by Research Councils to HEIs in support of the Government’s (and RCUK’s) policy preference for Gold OA. Such financial support will enable Publication Funds to be created at all affected HEIs to allow them to administer their allocated funds and determine how best to meet the APCs to secure free to user Gold OA publication.

As a result of an initiative by RCUK, this more substantive approach has now taken the form of a block funding allocation to over 100 individual institutions for the period beyond April 2013. The details of which they announced on 8 November 2012, as explained in RCUK’s summary progress note at Annex C (see http://www.rcuk.ac.uk/media/news/2012news/Pages/121108.aspx). HEFCE also intend to consult widely on how research outputs submitted to any REF subsequent to 2014 will be available by OA. HEFCE propose to allow HEIs to use their QR funding, to the extent deemed appropriate by the HEI, to supplement RCUK’s direct contribution thereby supplementing the financial resources available to HEIs for their Publication Funds.

The value of APC awards for years 2013–14 and 20014–15 were announced as being £17 million and £20 million respectively. Funding for financial years 2015–16, 2016–17 and 2017–18 could not be indicated at this stage, because of Treasury’s constraint on the commitment of funds for the next Spending Review period; but also, the level of funding required will be partly dependent on developments in the research publications market, which will be subject to change.

We recognise that this re-engineering of free access to published research will take time. RCUK anticipate a five year transition period during which time the level of take up for Gold OA might be expected to rise from an estimated 45% to 75%. Research Councils have estimated that the total allocation that they are prepared to make for APCs over the five year period could be in excess of £100 million. This further demonstrates the strength of their and the Government’s commitment to what is a radical re-engineering of the research publications market designed to improve access to research findings, stimulate innovation and contribute to economic growth. RCUK propose to review implementation of the policy in 2014 to make any appropriate mid-course corrections within the Government’s overall strategic policy.

RCUK’s announcement that their level of funding for Gold OA was designed to cover the cost of some 45% of Research Council funded published research in 2013–14 was based on their realistic estimate of the rate of change and initial level of demand from authors for Gold OA. However, RCUK’s implementation policy statement immediately raised questions in the minds of publishers and Learned Societies as to what embargo period would apply for Green OA for the balance of up to 55% of research publicly funded by Research Councils? A short period of uncertainty followed. Your inquiry affords me, and RCUK, an opportunity to clarify the settled position which was in the process of being resolved.

Embargo Periods for Articles Published Under the Green OA Model

The Government’s over-arching policy position on embargo periods for Green OA was clearly stated in response to the Finch Report. As stated in the published response, included as Annex A to this letter, the Government stated that:

“The Government has listened carefully to what publishers, learned societies and the Finch Group collectively have had to say on this issue. We prefer the “gold” over the “green” model, especially where the research is taxpayer funded so the Government agrees with the sentiment expressed in the Finch Report. Embargo periods allowed by funding bodies for publishers should be short where publishers have chosen not to take up the preferred option of their receiving an Article Processing Charge (which provides payment in full for immediate publication by the “gold OA” route). Where APC funds are not available to the publisher or learned society, for the publication of publicly-funded research, then publishers could reasonably insist on a longer more equitable embargo period. This could be up to 12 months for science, technology and engineering publications and longer for publications in those disciplines which require more time to secure payback. Even so, publications with embargo periods longer than two years may find it difficult to argue that they are also serving the public interest. “

The Government stands by this policy statement as published on Gov. Uk ( see https://www.gov.uk/government/news/government-to-open-up-publicly-funded-research )which also includes reference to the Decision Tree published on the Publishers Association website, with endorsement by BIS and RCUK in August 2012 (included in this response as Annex D).

RCUK accept, as stated clearly in their own policy statement (see http://www.rcuk.ac.uk/research/Pages/outputs.aspx) that Green OA is a legitimate second best alternative to Gold OA. RCUK also believe that implementation of their published OA policy should be consistent with the Government’s over arching OA policy position. In implementing their OA policy RCUK have been understandably concerned to ensure that every effort should first be made by researchers to operate within RCUK’s required publishing arrangements. This entails a preference for Gold OA and the use of short embargo periods19 where Green OA is being offered by publishers and Learned Societies. In some circumstances, publication on the basis of either of these two options may not be feasible. RCUK recognises that we are on a journey. Some researchers, particularly in certain arts and humanity disciplines, may only find it possible to publish in journals that because of the traditions of the discipline or its business model, or the availability of funding for Gold OA, may need a longer embargo period. RCUK recognises that the Government’s policy on embargo periods allows some discretion for this.

Government and RCUK will, for good reason, not want to weaken the Government’s policy preference for Gold OA, which is closely associated with CC-BY licensing20 conditions, or the alternative of Green OA with embargo periods of either six or twelve months depending upon the research discipline/journal, but where either option is judged by the HEI to not be feasible in the case of a particular research paper, the decision tree published in August 2012 by the Publishers Association and endorsed by BIS will apply. Furthermore, there may even be some cases where a longer embargo period than the 24 months indicated on the Publishers Association’s site may be needed and again the Government’s stated policy position above allows for this, but, such cases should be the exception rather than the rule. BIS’ understanding of the type of agreements on embargo periods being reached by the World Bank with publishers, on an individual journal basis, would reflect this expectation as well.

Engagement with Publishers, Universities, Learned Societies and other Stakeholders in the Development of the Research Council OA Implementation Policies

RCUK are leading on the development of their implementation policies for OA although BIS officials have engaged with them throughout. Further details are outlined in Annex C.

In addition to the RCUK-led implementation discussions, I am holding my own meeting with stakeholders. We have sought throughout to work with the publishing and research communities on our approach to open access, and success depends on all relevant parties believing the overall approach is workable. The Government’s Open Access policy for published research is simple in concept—namely that taxpayers should have access to the research findings they have funded and in a reasonable timescale, but it is undoubtedly complicated to implement. RCUK fully appreciate the fact that this is a journey not an event. The guidance they are providing will need to take account of issues as they arise. All participants in the process will need to accept that compromises will be needed from all sides.

Challenges and Concerns Raised by Stakeholders and How these have been Addressed

The development and implementation of a radically different OA policy intended to create benefits for all users (researchers, business including SMEs and the public) is a complicated and challenging task. The Finch Group, on which the full range of stakeholders were represented, cultivated an atmosphere of co-operation, sometimes out of initially conflicting positions. By creating a willingness to compromise, the Finch Group was able to produce a set of recommendations which (with the exception of the treatment of VAT for electronic publications which is an EU matter) the Government felt able to accept in full as stated at Annex A.

At the time of the Finch Group’s deliberations there was a tension between funders and publishers on the treatment or duration of embargo periods under the Green OA model. It was for this reason that the Government’s over-arching strategic policy position for Green OA embargo periods was adopted, which accommodates a range of circumstances whilst making the direction of travel clear. RCUK have pointed out that this will involve a significant transition period of some five years.

Other concerns have included the argument that the cost of Gold OA reduces research budgets. Government recognised this opportunity cost when formulating its policy position and its preference for Gold OA and still believes that the potential benefits that Gold OA affords (under the CC-BY licence conditions directly associated with it) outweighs the opportunity cost of about one% of the research budget. There is also positive evidence of universities now moving on to arrange the detailed planning for the new arrangements (see for example http://www.openaccess.cam.ac.uk ).

There is as yet limited empirical evidence of the quantifiable impact of Open Access on economic growth. Even so, the economic impact could be significant. Battelle’s recent report21 showed that between 1988 and 2010 genome sequencing projects, associated research and industry activity, both directly and indirectly, generated an economic (output) impact of $796 billion and created 3.8 million job-years of employment (310,000 jobs in 2010), with personal income exceeding $244 billion. It has been suggested that the economic success of the Human Genome Programme was partly attributable to the adoption of open access for the public programme of research.22

Concerns have also been raised that overseas based publications may choose to not adopt a similar OA policy to the UK Government. The concern is that if some leading overseas based journals do not choose to comply with the UK’s OA policy, UK researchers could, as a result, be denied publication in them with implications for the perceived credibility and importance of UK research. Whilst it is possible to have some sympathy with this concern, the Government needs to balance it against the wider policy objective to make publicly funded published research available in reasonable time, and preferably immediately, to UK taxpayers. Furthermore, publishers that specialise in OA journals appear to be increasingly attracting researchers and establishing their international credibility. There is also evidence of the global growth of Open Access (OA) for scholarly publications.

A most recent study of global trends has been by Mikael Laakso and Bo-Christer Bjork.23 This most current analysis of globally published scientific articles indicates that in 2011 some 340,000 scientific articles were published by 6,713 full immediate OA journals, of which some 49% required Article Processing Charges. Hence, OA represents some 17% of the 1.66 million articles published during 2011 according to Scopus, considered to be one of the most comprehensive article-level indexes of scholarly articles. Within this 17%, some 11% were immediately available (“Gold OA”) and the balance of 6% publicly available within 12 months of publication (“Green OA”). Hence, the UK is not as isolated as some have argued. The UK appears to be leading but going with the grain of the academic community and increasingly the publishing world. The number of OA published scholarly articles has grown, according to the same analysis, from a base of only some 20,000 OA scientific articles a decade ago. This is attributed to the growth of commercial publishers becoming key players in Open Access.

Biomedicine is estimated to have experienced a particularly rapid 16-fold growth in OA published articles between 2000 and 2011 but in terms of disciplines overall, the social sciences, arts and humanities (at about 56,000 OA articles in 2011) exceeds the number of OA publications in chemistry, chemical engineering, physics and astronomy combined (although it needs to be noted that there is supporting infrastructure for parallel forms of publication in some of these disciplines, such as physics).

Some Learned Societies have also expressed a lack of trust in HEIs to administer, in an equitable manner, the Publication Funds that they will receive from RCUK (and HEFCE). They are concerned that choices may be dominated by the interests of certain departments or researchers. This is an issue for HEIs to respond to, but the Government has seen no evidence to suggest that HEIs will act irresponsibly. It is important for them to have autonomy and flexibility in such decisions so that they may use their funding24 most efficiently and in a way that best suits their publication priorities and have the capacity to stimulate further competition and innovation in the research publications market. The most efficient and competitive publishers will rise to this challenge and new entrants may also appear.

There is also the concern amongst HEIs and some publishers that rather than securing a first mover advantage the UK stands to lose out to the “free-rider” problem, meaning that the UK pays and others benefit. The Government is fully aware of this risk, but as for all first movers it is a risk that has to be accepted. In the worst case, the UK will still benefit from ensuring that eventually 75% of all Research Council funded published research will be immediately available for free to users. This in itself is a prize worth having. In addition to which, in the same manner that the UK is communicating globally the fruits of its research through the recently released beta test version of the Gateway to Research (see http://www.rcuk.ac.uk/research/Pages/gtr.aspx), access to UK research will demonstrate the strength of the UK research base and further encourage international collaboration with the UK and inward investment to our long-term advantage. It is also expected be to the advantage of UK based researchers in terms of their citation rates, increasing their global recognition.

We are not being complacent about the “first-mover” initiative we have taken and the risk of doing so. Discussions have taken place with the EU and the Commission which has now stated that it will expect all research published under Horizon 2020 to be on an OA basis. The UK have impressed on them the relative merits of Gold OA. The Commission will make open access to scientific publications a general principle of Horizon 2020, the EU’s Research & Innovation funding programme for 2014–20. As announced in July 2012, immediately following publication of the UK’s OA policy, as of 2014, all articles produced with funding from Horizon 2020 will have to be accessible on an Open Access basis:

articles will either immediately be made accessible online by the publisher (“Gold” open access)—up-front publication costs can be eligible for reimbursement by the European Commission; or

researchers will make their articles available through an open access repository no later than six months (12 months for articles in the fields of social sciences and humanities) after publication (“Green” open access).

For Green OA under the Commission’s requirements, it is not yet clear how consistently repositories would be used and what version of the paper would be published on them, that is whether it would be the finally published paper in pdf format or the author’s manuscript copy (that remains the author’s copyright and which can be deposited immediately on ArXiV, Institutional repositorities or EuroPubMed now without any restriction.)

The Commission has also recommended that Member States take a similar approach to the results of research funded under their own domestic programmes. The goal is for 60% of European publicly-funded research articles to be available under Open Access by 2016. That coincides with the level of penetration assumed by RCUK for the UK in 2015–16.

(see http://europa.eu/rapid/press-release_IP-12–790_en.htm).

The EU policy position, which is a recommendation and not a legislative proposal, includes the UK’s preferred Gold OA. For Green OA the Commission is consistent with RCUK’s policy objective. Even so we will maintain with the Commission particularly during the transitional five year period, that where there is no funding available to pay APCs for Gold OA the required embargo periods for Green OA should be 12/24 months as allowed for by the UK not six–12 months as currently being required by the Commission.

Similarly, attitudes towards the adoption of OA are becoming more favourable in the US. The National Institute of Health (NIH) has recently declared that it will with hold grant awards if research is not published in compliance with its existing OA policy which requires OA publication on Pub Med Central http://www.ncbi.nlm.nih.gov/pmc within 12 months (see http://grants.nih.gov/grants/guide/notice-files/NOT-OD-13–016.html ).

The UK Medical Research Council and the Wellcome Trust have operated a similar repository arrangement for medical research since 2007 requiring publication within six months. In July 2011, the European Research Council became the third European funder to join UKPMC, following Telethon Italy and the Austrian Research Fund. As a result of this participation, the 18 existing UK and European funders agreed that the UKPMC service should be rebranded as Europe PMC which took place on 1 November 2012 further reinforcing the UK Government’s policies on transparency and open access. Information on Europe PMC can be found at the new website www.europepmc.org, details of the NIHR’s OA policy is available at http://www.nihr.ac.uk/research/Pages/Research_Open_Access_Policy_Statement.aspx

Europe appears to account for some 37% of global OA articles and N. America for about 19%, but Asia is also responsible for some 25% of global OA, and as stated above the use of OA is increasing and the proportion of Gold OA growing with it. Hence, the fear that the UK may be acting alone is probably unfounded.

The publishers and Learned Societies are a highly valued part of the fabric of the UK’s world class research base. Their willingness to collaborate through the Finch Group and to compromise on some of their aspirations (by accepting, for example, the need to offer CC-BY conditions as part of the Gold OA APC transaction) whilst signalling through their public library initiative their desire to contribute to the public good suggests that there is a real prospect of the UK implementing an OA policy, based on the appropriate use of Gold and Green, in a way that allows all stakeholders to prosper.

The Public Library initiative (PLI) , proposed by the Publishers Association during the course of the Finch Group’s deliberations is an example of how the publishers and Learned Societies contribute to the public good. The PLI represents a valuable concession by publishers to facilitate free walk-in access to all of their research publications in public libraries. This would apply to globally sourced research and not just the 6% of global research that is sourced from the UK (see Annex E). The Government now looks forward to this PA initiative being implemented in 2013.

I look forward to discussing the issue with you on 29th January.

THE RT HON DAVID WILLETTS MP

Annex A

LETTER TO DAME JANET FINCH ON THE GOVERNMENT RESPONSE TO THE FINCH GROUP REPORT: “ACCESSIBILITY, SUSTAINABILITY, EXCELLENCE: HOW TO EXPAND ACCESS TO RESEARCH PUBLICATIONS”

Dear Janet,

Please find enclosed the Government’s response to your excellent report on open access to published research.

We are firmly committed to improving access so the Government accepts the proposals in your report, except for one specific point on VAT. Reference was made to the issue of VAT being applied to e-journals but not printed books and journals. Consideration has been given to this, but, in consultation with Treasury it has become evident that current VAT rules agreed at EU level preclude a reduced or zero rate for e-journals. The enclosed note sets out our response in more detail.

I welcome the OA policies being announced by funding bodies. I also welcome the publishers’ proposed initiatives for improving access for SMEs and for the public libraries.

You have suggested that your Group should reconvene in a year’s time to reflect on progress. That is an excellent idea. I would be pleased to join you for that meeting to review progress made with implementing your excellent report.

Thank you for all you have done to help us open up our world-class research base to more people, which will no doubt benefit all of the UK. Please pass on my thanks to the other members of your group and Michael Jubb (Director RIN) who supported you.

David Willetts

GOVERNMENT RESPONSE TO THE FINCH GROUP REPORT: “ACCESSIBILITY, SUSTAINABILITY, EXCELLENCE: HOW TO EXPAND ACCESS TO RESEARCH PUBLICATIONS”

We are grateful to the Finch group for their constructive investigation and we welcome the report. We wish to extend open access to research and so accept all the conclusions in the report (except for one involving tax which will be considered further).

Taking each of the Group’s recommendations in turn:-

(i)a clear policy direction should be set towards support for publication in open access or hybrid journals, funded by APCs, as the main vehicle for the publication of research, especially when it is publicly funded;

(ii)the Research Councils and other public sector bodies funding research in the UK should establish more effective and flexible arrangements to meet the costs of publishing in open access and hybrid journals;

The Government agrees with both of these recommendations. We recognise that whilst Open Access (OA) means free access to the user and full right of search, it does not follow that OA has no cost. Support for publicly funded research institutions will be needed to pay the cost of APCs this funding will come out of existing research funds.

Funding bodies are now in the process of publishing their respective policy positions on OA, which will include provision for the funding of APCs. Details of the precise funding mechanisms will be included in their respective announcements though they are also co-operating with each other to ensure a co-ordinated approach.

RCUK currently provides support for APCs through both direct and indirect costs as part of grant funding. They will now put in place a simpler, more flexible and transparent mechanism that will allow institutions to set up publication funds to cover such charges. The most suitable payment mechanism is currently under discussion and includes options such as direct cost within a grant application (either open to applicants to make a case or through a formula) or as a block grant to universities.

In all cases universities upon receipt of funding should transfer these charges to their institutional publication fund. A university can then use these funds to pay for APCs for any article resulting from research council funding. Research Councils will monitor compliance with its policies at grant level through its outputs systems. Once Research Councils have established the payment mechanism, operational details will be set in discussions with the academic community.

(iii)support for open access publication should be accompanied by policies to minimise restrictions on the rights of use and re-use, especially for non-commercial purposes, and on the ability to use the latest tools and services to organise and manipulate text and other content;

The Government welcomes this recommendation which is consistent with our aspirations for implementation of the Hargreaves recommendations. Where APCs are paid to publishers, the Government would expect to see unrestricted access and use of the subject content and the details of how this should be best achieved will be addressed in the detailed policy statements to be published by funding bodies.

In relation to Hargreaves, the Government believes any exception on copyright—on which decisions are yet to be taken—must be compatible with the broad approach of the Finch report.

(iv)during the period of transition to open access publishing worldwide, in order to maximise access in the HE and health sectors to journals and articles produced by authors in the UK and from across the world that are not accessible on open access terms, funds should be found to extend and rationalise current licences to cover all the institutions in those sectors;

The Government understands and supports the objectives behind this recommendation. The extent to which funds can be made available for this purpose will be a matter for the independent funding bodies.

(v)the current discussions on how to implement the proposal for walk-in access to the majority of journals to be provided in public libraries across the UK should be pursued with vigour, along with an effective publicity and marketing campaign;

The Government welcomes this imaginative and valuable initiative by the publishing industry. We encourage the working group that has already been set-up to address it, which includes public library representation, to press ahead and implement the proposed two-year pilot scheme at the earliest opportunity. We hope it will become a permanent feature of the UK’s public library service.

(vi)representative bodies for key sectors including central and local Government, voluntary organisations, and business should work together with publishers, learned societies, libraries and others with relevant expertise to consider the terms and costs of licences to provide access to a broad range of relevant content for the benefit of consortia of organisations within their sectors; and how such licences might be funded;

(vii)future discussions and negotiations between universities and publishers (including learned societies) on the pricing of big deals and other subscriptions should take into account the financial implications of the shift to publication in open access and hybrid journals, of extensions to licensing, and the resultant changes in revenues provided to publishers;

The Government encourage the various stakeholders to pursue these two recommendations. We look to JISC to contribute its long-standing experience in this field to help in such negotiations and particularly with regard to implementing at the earliest opportunity the proposed extension of licensing to high-technology Small and Medium Sized Enterprises (SMEs).

(viii)universities, funders, publishers, and learned societies should continue to work together to promote further experimentation in open access publishing for scholarly monographs;

The Government welcomes this recommendation.

(ix)the infrastructure of subject and institutional repositories should be developed so that they play a valuable role complementary to formal publishing, particularly in providing access to research data and to grey literature, and in digital preservation.

The UK Research Councils have already invested in a number of successful repositories. Notable examples include the Economic and Social Research Council’s Research Catalogue (see http://www.esrc.ac.uk/impacts-and-findings/research-catalogue/index.aspx) and UKPubMed which has been funded by the Medical Research Council (MRC) and the Biotechnology and Biological Sciences Research Council (BBSRC), the Chief Scientist Office, part of the Scottish Government Health and Social Care Directorates and other funding bodies (see http://ukpmc.ac.uk/About). Indeed, in 2010 just over 40% of the articles published that year, or almost 70,000 articles in real numbers, were Open Access (OA) see http://ukpmc.blogspot.co.uk/2012/05/increasing-proportion-of-ukpmc-articles.html.

The Government has recently provided £150 million for the development of e-infrastructure that should benefit these OA objectives. It has also committed £75 million to the development of the ELIXIR project at the European Bioinformatics Institute, Hinxton, to create a world-leading repository in bioinformatics but, generally, the development of infrastructure for subject and institutional repositories will primarily be a matter for institutions themselves. Even so, the Government wish to ensure that the UK secures the greatest added value from such developments. The “Gateway to Research” being developed by RCUK will provide an additional tool to sign and facilitate public access to the full body of research funded by the UK Research Councils by late 2013.

(x)funders’ limitations on the length of embargo periods, and on any other restrictions on access to content not published on open access terms, should be considered carefully, to avoid undue risk to valuable journals that are not funded in the main by APCs. Rules should be kept under review in the light of the available evidence as to their likely impact on such journals.

The Government has listened carefully to what publishers, learned societies and the Finch Group collectively have had to say on this issue. We prefer the “gold” over the “green” model, especially where the research is taxpayer funded so the Government agrees with the sentiment expressed in the Finch Report. Embargo periods allowed by funding bodies for publishers should be short where publishers have chosen not to take up the preferred option of their receiving an Article Processing Charge (which provides payment in full for immediate publication by the “gold OA” route). Where APC funds are not available to the publisher or learned society, for the publication of publicly-funded research, then publishers could reasonably insist on a longer more equitable embargo period. This could be up to 12 months for science, technology and engineering publications and longer for publications in those disciplines which require more time to secure payback. Even so, publications with embargo periods longer than two years may find it difficult to argue that they are also serving the public interest.

BIS 16 July 2012

Annex B

RCUK ALLOCATION OF THE £10 MILLION FOR THE INITIAL INTRODUCTION OF “GOLD OA”

1. RCUK’s policy on Access to Research Outputs is available at:

www.rcuk.ac.uk/research/Pages/outputs.aspx

2. The institutions that will receive the additional funding are:

Cardiff University
Durham University
Imperial College London
King’s College London
Loughborough University
Newcastle University
Queen Mary, University of London
Queen’s University of Belfast
The University of Manchester
University College London
University of Aberdeen
University of Birmingham
University of Bristol
University of Cambridge
University of Dundee
University of Edinburgh
University of Exeter
University of Glasgow
University of Leeds
University of Leicester
University of Liverpool
University of Nottingham
University of Oxford
University of Reading
University of Sheffield
University of Southampton
University of St Andrews
University of Strathclyde
University of Warwick
University of York

Annex C

UPDATE ON IMPLEMENTATION OF RCUK POLICY ON OPEN ACCESS, DECEMBER 2012.

The RCUK Policy on Open Access

The RCUK Policy on Open Access applies to peer-reviewed research papers that would normally be published in journals or conference proceedings, and which result from research acknowledges funding from the Research Councils. It does not apply to other forms of scholarly output, such as books or monographs. The policy builds on individual councils’ policies on open access which have been in place since 2005/6.

Peer-reviewed research papers submitted for publication from 1 April 2013:

be published in journals which are compliant with Research Council policy on Open Access, and;

Must include details of the funding that supported the research, and a statement on how the underlying research materials such as data, samples or models can be accessed.

To be compliant with the policy, Journals must offer either a “Gold” Open Access option, which results in immediate and unrestricted access to the published version of a paper via the journal’s web site; or a “Green” option, allowing deposit of the authors final peer-reviewed manuscript in an institutional or subject-specific repository, with a maximum embargo period on access of six months (or 12 months for AHRC & ESRC funded research). In addition, the policy requires publishers to use the Creative Commons “Attribution” licence (CC-BY), when an Article Processing Charge (APC) is levied.

Summary of the main RCUK actions since publication of the Finch Report (June 2012):

16 July: Revised RCUK Policy on Open Access launched, alongside the Government response to the Finch report, to come into force on 1 April 2013.

7 September: Announcement of additional £10 Million provided by BIS to pump-prime activities in Open Access during FY 2012–13.

8 November: Announcement of RCUK funding mechanism to support payment of Article Processing Charges for “Gold” OA, and the length of the transition period.

13 November: RCUK workshop for key HEIs on implementation of RCUK OA policy.

6 December: Russell Group convened meeting with publishers, learned societies and RCUK to discuss the RCUK OA policy.

Reaction to the RCUK OA policy

Much of the reaction to the revised policy has been supportive and the main concerns are not with the policy as such but its implementation and the speed of the transition to a new “normal” of Open Access.

Some learned societies in the Humanities, Arts and Social Sciences (HASS) have expressed concerns about making the Gold OA model work for their journals, because of specific issues over the small number of authors and the longer average article length, in relation to the number of subscribers, which would result in what are considered to be unsustainably large APCs. There is also considerable concern in the HASS community that HEIs will spend the RCUK block grant disproportionately on articles in the natural sciences. RCUK has no evidence that this will be the case. Whether it is will be part of the review in 2014.

In addition, there have been objections from some in the HASS community to the requirement to use the CC-BY licence for papers published using the Gold OA model. Some of the issues (for example, use of 3rd party material in papers) need further exploration. However, much of the discussion seems to be based on misinterpretation and misrepresentation of what the CC-BY licence will and will not allow. RCUK has concerns that some communities are proposing to use the CC-BY-NC licence, which by disallowing commercial re-use, will impose barriers to the full re-use of published papers, and stifle innovation within scholarly publishing.

Some of the feedback from HEI community gives the impression that they are expecting RCUK to provide all the solutions, rather than recognising that HEIs have a key role to play. For example, by just stating that RCUK is not providing sufficient funding, rather than acknowledging that the HEI community must play a key role in negotiating with publishers to drive down subscription and APC costs to enable the available funding to go further.

There is also a continuing and vocal campaign by a number of OA “pioneers” who consider that RCUK and Finch have made a major mistake in supporting the Gold approach to OA, in preference to Green. Their concerns are based around cost (Gold is using money that could be spent on research) and, from their perspective, the lack of any demonstrated requirement for re-use requiring a CC-BY licence. The RCUK position is that disseminating research is just as much a cost of research as is hiring researchers, buying consumables, and so on.

Funding for Gold Open Access

The total amount of funding that RCUK will put into supporting Gold open access is based on estimates of the numbers of publications arising from all Research Council funded research activities. Publications data from 2010 & 2011 indicate that some 26,000 peer-reviewed research papers per year arise from Research Council funded research, of which approximately 90% are produced within the HEI sector and 10% from Research Council institutes. The average cost of an APC has been taken from the Finch report (estimated as £1,727 plus VAT). The total value of the RCUK APC fund for supporting the HEI sector has therefore been set as follows:

Year	2013–14	2014–15	2015–16	2016–17	2017–18
Value of RCUK APC fund	£17M	£20M	£TBD	£TBD	£TBD
Estimated % papers block grant will fund as Gold OA	45%	53%	60%	67%	75%

The value of APC awards from 2015–16 onwards are indicative only, and will depend on the outcome of the next Spending Review, and any changes in funding resulting in the review of the policy planned for 2014.

This increase in funding over a five-year transition is a reflection of the time that will be needed for researchers, institutions and publishers to transition into a “Gold” OA model, as well as to allow existing publication funding already provided through direct and indirect costs as part of grant funding to be fully utilised. RCUK estimates that the steady-state will be that some three-quarters of research council funded research will be published using the “Gold” OA route and a quarter using “Green”. However, we will review this figure during the transition phase as we gather further publication data.

Universities will receive block grants in proportion to the amount of direct labour costs (“directly-incurred” and “directly-allocated”) awarded on grants provided it is above a threshold total of £10,000, that they have received over the three years from April 2009 to March 2012. Direct labour costs have been used as a proxy of research effort leading to the generation of publications, independent of the effects of equipment and infrastructure costs, and overheads. The 36 institutions from the Russell and 1994 groups of research intensive HEIs will receive just over 80% of RCUK’s APC funding. Seventy-one other institutions share the remaining funding. In addition, the top 30 HEIs (in terms of total funding from the research and funding councils) have shared in the distribution of the £10M made available by BIS to help kick-start OA developments (to be spent by 31 March 2013).

SHARE OF THE RCUK FUNDING ALLOCATED TO THE 36 RUSSELL/1994 GROUP HEIS

RCUK Activities to Support Policy Implementation

Facilitating a project within the HEI sector to develop common processes and procedures, and to share best practice to support implementation of the RCUK policy. RCUK will provide funding to support a project manager and discussions are underway with a number of organisations about hosting the project office.

In conjunction with the Wellcome Trust, commissioning the SHERPA-RoMEO group at Nottingham University to develop a support service to provide easily accessible and understandable advice on how journals provide compliance with the RCUK and WT policies.

In conjunction with the Wellcome Trust, we have written to the top-60 publishers of Trust and RCUK funded research to inform them of the changes in policy and to ask how they plan to comply with policy. This process has resulted in constructive dialogue with the major publishers, and with publisher trade bodies (STM, ALPSP, PA & OASPA). Meetings with the trade bodies are continuing on a regular basis.

RCUK will be holding a series of bilateral meetings with key stakeholders between now and mid-February, to discuss details of implementation of the policy. These include a meeting with the Russell Group, with a range of Learned Societies and with publishers—in The Arts, Humanities & Social Sciences; Biomedical sciences; and “other” Science & Technology. We expect to recognise the outcomes of these meetings in the final guidance document to be published in late February. We are aware that the implementation issues are significant.

It is planned to issue updated guidance on implementation of the policy in mid-February, following this process—current guidance is on the RCUK website. We recognise that the funding we are providing to HEIs to implement the policy is based on a number of estimates. We will therefore undertake a review of the policy and its implementation in 2014. We are making it clear that if there is real evidence that the policy is not working, is producing unintended consequences, or that the level of funding we are making available to support it is insufficient, we will take this evidence very seriously as part of the review. However, at present we are of course unable to make any commitments beyond the current SR period.

Sustainability

A common criticism is that the Research Councils are diverting money that would otherwise be spent on research into paying additional money to publishers in Gold APCs. The Finch Report recommended that “Gold” is the only long-term sustainable solution for publishers. However, RCUK considers that sustainability cuts two ways. The model also needs to be sustainable for funders and institutions. RCUK will make the data they collect on APC fees paid through their block-grant mechanism publicly available. We expect HEIs, JISC Collections and others (eg RLUK) to negotiate hard with publishers to drive down subscription charges to reflect the additional funding that they are receiving to support the payment of APCs. Ideally, we would expect publishers to introduce differential pricing in the UK market to reflect additional income they are earning through APCs.

Compliance Monitoring

The Research Councils recognise that implementation of its policy on Open Access will require a major change in the way researchers, institutions and publishers manage the process of publishing the results of the research that we fund. The Research Councils are also asking that this cultural shift takes place over a relatively short period of five years. For these reasons, the Research Councils see this transition to full Open Access as a journey and not as a single event.

During the transition period we expect researchers and their institutions to follow the spirit of the policy and strive to achieve full compliance. As the available funding for Gold Open Access increases during the transition phase, so will our expectations of compliance. At the end of the transition period we will expect researchers and institutions to be fully compliant with the policy, and for 100% of research papers then arising from the research we fund to be published in journals which are compliant with our policy on Open Access.

In the early years, rather than wielding a big “compliance stick”, we want to work with the HEI sector to focus on changing the way researchers publish their research, to help facilitate a sustainable transition to a new “normal” of Open Access.

RCUK

17 December 2012

Annex D

DECISION TREE FOR THE DETERMINATION OF EMBARGO PERIODS

This flow diagram below was drawn up by the Publishers Association in consultation with BIS and RCUK and is on the Publishers’ Association website and available from the Gov.UK website.

http://www.bis.gov.uk/news/topstories/2012/Jul/government-to-open-up-publicly-funded-research and also directly on the Publishers Association site at:

http://www.publishers.org.uk/index.php?option=com_content&view=article&id=2299:finch-willetts-rcuk-green-oa-and-embargoes&catid=503:pa-press-releases-and-comments&Itemid=1618

Annex E

JOURNAL PUBLISHERS CONSIDER FREE WALK-IN ACCESS VIA PUBLIC LIBRARIES

2 May 2012

Scholarly journal publishers collectively are exploring fee-waived walk-in access via the public library network, as part of a package of proposals under discussion in the Finch Working Group on extending access for UK researchers to global published research findings. A number of technical and licensing issues are still under review by a specialist publisher-library group, such as authentication, copying rights, and network security, but the expectation is that these matters can be resolved within an effective and sustainable overall package of recommendations from the Finch process. The repertoire available would be assembled and licensed on an opt-in basis, probably via the Publishers Licensing Society.

David Willetts MP, Minister for Universities and Science, speaking at the AGM of The Publishers Association said: “Open access is the way forward and this proposed PA initiative would be a very useful way of extending public access to the majority of articles currently only available to public libraries through subscription. If agreement can be reached on the PA’s related conditions, it would be good for our libraries too.”

Steven Hall, Managing Director of Institute of Physics Publishing said: ”This is an important initiative to improve access beyond the research libraries and institutions with direct subscriptions. It must be seen however as an element in a balanced package in which evolving effective funding mechanisms to support gold open access and agreeing acceptable embargo periods for green open access mandates will be the critical issues, alongside extending existing licensing arrangements sector-by-sector through separate negotiations.”

Graham Taylor, Director of Academic Publishing at The Publishers Association said: “We expect this significant initiative will be a welcome new means of extending access to journal articles for SMEs, entrepreneurs and citizen scholars who might not have ready access to a research library. The repertoire of primary global research has hitherto not been available via the public library network.”

Notes for Editors

1. Finch is expected to report in June. The independent Working Group on Expanding Access to Published Research Findings chaired by Dame Janet Finch was established with DBIS support to investigate how access to global research outputs could be extended for the benefit of UK researchers. The WG has been operating since September 2011 and includes representatives from research libraries, learned societies, the funding councils, the research community, the universities and society, commercial and open access publishers.

2. This initiative is confined to primary research journals only. It does not extend to textbooks, research monographs (books), or aggregated databases already available to public libraries, nor would this offer extend to the consumer trade sector in any way.

3. For further information, contact Graham Taylor (gtaylor@publishers.org.uk This e-mail address is being protected from spambots. You need JavaScript enabled to view it) or Gemma Hersh (ghersh@publishers.org.uk This e-mail address is being protected from spambots. You need JavaScript enabled to view it) at The Publishers Association, or Emma Griffiths at BIS (emma.griffiths@bis.gsi.gov.uk This e-mail address is being protected from spambots. You need JavaScript enabled to view it)

The Publishers Association

The Publishers Association (PA) is the leading trade organisation serving book, journal and electronic publishers in the UK. Membership is comprised of 120 companies from across the trade, academic and education sectors. Its core service is representation and lobbying, around copyright, rights and other matters relevant to members, who represent roughly 80% of the industry by turnover. www.publishers.org.uk

APPENDIX 2

ECONOMIC ANALYSIS OF COST OF ALTERNATIVE OPTIONS FOR OPEN ACCESS IN THE UK

Summary of Modelling Assumptions and Findings

Background

This analysis was carried out by BIS to provide an impartial assessment of the potential cost of open access (OA) policies and to inform policy advice for BIS’ response to the recommendations of the Finch Group.

The analysis was undertaken prior to the recommendations of the Finch Group and the RCUK policy announcements.

The analysis is therefore not an evaluation of the policy subsequently adopted by RCUK but an assessment of:

1.The state of open access to scholarly research in the UK (as per information in the Summer of 2011, when the analysis presented in this paper commenced).

2.Two alternative methods for increasing the take up of open access in the UK evaluated on the basis of their cost effectiveness with respect to the current state of OA.

This analysis considers the costs of open access, but not the economic and other benefits that would occur as a result of increase access to scholarly research. Nevertheless, open access holds potential for enabling open innovation and speeding up the rate of technical progress that underpins sustainable long term growth. To support open innovation, the Government has a role in facilitating knowledge flows between the market and non-market sector. As explained in the BIS Innovation and Research Strategy for Growth (2011), the Government can provide an infrastructure that allows users to exploit the findings of research more swiftly and a suitable open access policy is instrumental in this role.

Evidence on the quantity and quality of open access to scholarly research in the world and in the UK is incipient and scarce. For this reason all analyses on this matter rely on certain necessary assumptions. For the projections in this paper, the starting point is the year 2011, which is derived by applying to publication counts prevailing in 2010 the observed growth rate during 2005–10 and assuming that this growth is approximately constant thereafter. This assumption of constant growth ensures that all changes in cost observed are due to the difference in growth rates of OA and non-OA publications in the UK and elsewhere. Both for simplicity and due to a lack of reliable information, the estimates are not adjusted to account for future price inflation either nationally or in the market for scholarly research.

The analysis compared alternative possibilities for expanding access without reference to a preferred policy option. The comparative analysis is therefore based on projections, not predictions. Unlike a prediction, a projection estimates the possible evolution of open access in the UK and elsewhere on the basis of some stated assumptions. Because not enough is known about the way in which open access expands, BIS assumed the simplest model possible: constant growth departing from a known initial condition, the year 2010. In addition, assumptions on the unit costs of open access were made using actual information available at the time so as to approximate the total national costs of the projected open access policies. All assumptions are clearly stated below but they need not hold in an uncertain future. These projections are valid for the purposes of comparing alternative options because the same assumptions apply to all the alternatives, but should not be regarded as reliable predictions of the future evolution of open access. They do, however, represent the best comparative projections BIS could produce given the information available in the summer of 2011.

Following from the Finch Group’s working definition, open access is the ability to download, read and print electronically-published refereed journal-articles, leaving aside research content that is accessible in other formats. Text mining is an additional service that may or may not be offered together with downloading, viewing and printing. The presence of the additional service of text mining increases the cost of open access.25

In the UK and elsewhere open access (even without text mining) is available only for a subset of published articles. Even so, open access is expanding worldwide at a fast pace.26 There are many reasons why the findings of research are not immediately available upon publication. The Finch report gives a detailed account of the multiple interests and pressures that come together in the market for scholarly publishing.

The Market for Scholarly Publishing

BIS economic analysis of options considers only pricing in the market for scholarly publishing: whether a price should be paid for accessing a refereed journal article online and if a price was justified what determines the level at which this price is set in the market for scholarly publishing. It is worth bearing in mind that in any market price setters may well set a price of zero for access, and the evidence on scholarly publishing is that some open access publishers indeed do. Because the market for scholarly publishing is not price-regulated, BIS took existing business models and price schedules as given in order to calculate the nation-wide cost of alternative policy options.

The main reason why any supplier charges a price is to recoup costs of production. This is no different in scholarly publishing. A refereed journal article is a bundled commodity made up of three elements: a research manuscript, a referee process over the manuscript and the publisher’s certification that the referee process has taken place. The costs of the manuscript are borne by researchers and their funders. The costs of refereeing are shared between researchers (and their funders) and publishers who organise the refereeing. The costs of certification, that is formatting the article to the journal specification, are largely borne by publishers.27 Payments for open access, called Article Processing Charges (APCs), are for recouping the cost of certification, uploading and maintaining the electronic version of the refereed journal article—all of which fall on the publisher who is thus the price-setter for APCs.

This analysis is not concerned with determining a clearing price for APCs, which, even if possible, would effectively imply regulating the market for scholarly publishing, which in turn is a global not domestic market. APCs can be expected to respond to business models and needs domestically in the UK and worldwide and possibly vary by research discipline and by the perceived standing or status of the journal’s within any given research discipline. For the purpose of projecting the economic value of alternative policies, certain average APC prices have been assumed at levels explained below.

Property rights over the bundled commodity “refereed journal article” are also different for its components and this matters for open access policy options. Leaving referees aside, the manuscript belongs to the author(s), but the journal referee certification belongs to the publisher. This means there are at least two non-exclusive routes for accessing the findings of scholarly research:

1.the author exercises her property rights and deposits the manuscript (not the refereed article28) in an open electronic repository or digital library for free—known in this analysis as GREEN access

2.the publisher exercises its property rights and charges a price (that could be zero) for making the refereed journal article immediately accessible on publication. Two common options for the publishers to collect revenue for access are:

users pay a journal subscription for accessing articles that are otherwise embargoed for some period of time;29 and

authors pay an article processing charge (APC) for the publisher to make the article available to any user immediately on publication- known as GOLD access.

Note that even if hypothetically the whole of the UK’s refereed journal articles were available in open access, under gold or green, UK publications would still account for only 6% of the World output of refereed journal articles, based on figures for 2010. To access publications from the rest of the world which are not available in open access, the UK must pay the price set by publishers, most commonly in the form of subscriptions

The State of Open Access in Britain and the World

The proportion of OA-articles in the World was estimated to be around 10% in 2010 (Laakso 201130) and growing at annualised rates of 20% in the previous five years. If this rate of growth was sustained until 2020, open access would reach 42% world wide by that year.

In 2010 UK researchers published 123,600 articles, attracting a 6% share of World publications (1,936,000). Under BIS assumptions,31 the UK published 7% of world open access in 2010, which accounted for 11% of UK publications.

In Figure 1 growth rates are assumed constant to show that if trends prevailing prior to 2010 were sustained, 51% of all UK publications would be open access by 2020. Using this evolution of publications as the baseline it is possible to calculate the costs of maintaining the baseline trend.

There are two components of the total national cost of maintaining the baseline evolution: the costs of maintaining open access and the costs of subscribing to world articles that are not open access.

The costs of maintaining open access: BIS estimates that for the circa 16,000 open access articles published in 2011, on average UK researchers paid approximately £750 per article.32 If this unit cost was constant over time, the gross nominal cost of maintaining organically growing open access trends would raise to £25 million in 2015 and £63 million in 2020.

The cost of accessing articles that are not open access: Subscription costs are projected to fall because open access worldwide is growing faster than total publication counts. This means that a larger proportion of world articles is available under open access over time and thus subscriptions should not have to be paid for those. If open access were to grow constantly as projected and assumed subscription rates per article33 were constant, then the total costs of subscriptions would fall from £143 million in 2011 to £133 million in 2020.

Combining the two above factors into a single net effect, in the baseline scenario, if 2011 trends were maintained, the total UK APC and subscription cost of accessing scholarly research would rise from £155 million in 2011 to £175 million in 2015 and £196 million in 2020.

Adding all estimated expenses from 2011 to 2020 leads to a total cumulative cost of £1.8 billion over 10 years. A full breakdown of this and other calculations presented below is provided in Table 1.

Expanding access with GOLD open access

The current state of open access does not guarantee that all open access is text mineable and fully accessible on publication, as dictated by gold open access. One way of expanding access to scholarly research in the UK is to improve the accessibility of the organically growing open access publications so that articles projected to be open access under the current state are gold accessible.

Following the relevant literature (Ambruster, 2010)34 and the experience of the Welcome Trust, the average APC per article in gold open access was assumed to be £2,000 per article.

The costs of granting gold access to baseline open access: If current trends were constant over time, provision for charges for APC to guarantee gold access for the organically growing open access in the UK would have required £32 million hypothetically35 in 2011 to provide 13% of all UK articles in Gold open access. Under constant growth assumptions, this cost would be projected to rise to £67 million in 2015 and £167 million in 2020. These magnitudes compare to current spending estimates of £12 million, £25 million and £63 million respectively in each of those years.

The costs of accessing articles that are not open access would be the same as in the baseline because the proportions of open and non-open access are the same.

The additional costs of Gold over baseline are £42 million in 2015 and £104 million in 2020. This is the additional funding required, under the assumptions made in this paper, to guarantee fully minable open access to 51% of all (projected) publications attributable to the UK in 2020.

Adding together the estimated yearly costs of gold access plus subscriptions obtains a cumulative total bill of £2.3 billion over the 10 years from 2011 to 2020, which is £500 million in excess of the 10 year baseline projection presented above. Year-on-year projections are presented in Table 1.

Expanding Access through Research Councils

An alternative to granting gold access to organically growing open access is to add to the organically growing level by expanding the coverage of open access. The aspiration of having all articles attributable to Research Councils’ funding in gold access has the potential for expanding the coverage of open access beyond the current state in the UK. The benefit of this Research Councils’ gold option over the gold estimate above is that it implies more articles available in open access than is organically available currently. The average APC is maintained at £2,000 per article so that differences in the total costs are due to coverage not to price differences

Publications attributable to Research Councils account for over a quarter of the UK total, numbering 32,000 in 2010. This is nearly double the estimate of open access in the current state at 16,000. As a result, if all publications attributable to Research Councils were available in gold open access, this would take proportions of all UK publications available in OA to 43% in 2015 and 64% in 2020 (see Figure 2). Associated costs can be projected as follows:

The cost of granting gold open access to all publications attributable to Research Councils: making all articles attributable to Research Councils available in gold access at an average of £2,000 per article would cost a hypothetical36 £66 million in APCs 2011, rising to £74 million in 2015 and £86 million in 2020.

The cost of accessing articles that are not open access. Assuming the same subscription rate per article as in the baseline33 and given that more articles than in the baseline are now available in open access, subscription costs would fall from a hypothetical £140 million in 2011 to £132 million in 2020.

Combining the two factors above, Table 1 shows that the total net cumulative nominal costs of a Research Councils’ gold access projection from 2011 to 2020 adds to £2.2 billion over 10 years to deliver 64% of UK publications in open access. This is an additional £400 million over 10 years, above the baseline whereby it costs £1.8 billion to deliver only 51% of publications in open access by 2020.

Comparative Analysis

Disregarding the impossibility of reaching full compliance with either UK Gold access or Research Councils’ gold access as a one-off move in 2011, BIS’s estimates suggest that, on the basis of a linear growth model, the Research Councils’ gold option has the potential of delivering better value for public money than the alternative. This is because the Research Councils’ policy would deliver a step change in the level of open access, and thus be more expensive in the early years, but over time Research Councils’ publications grow more slowly than organically growing open access. The option of paying (even for a step change initially) for the slower growing publication stream bears lower deadweight cost because the faster growing open access publication stream occurs anyhow in the baseline scenario and would be financed from outside the policy.

Figure 3 demonstrates why Research Councils’ proposal has lower cumulative cost. The funder would be paying for Gold access for the Research Councils’ gold line, thereby increasing the total open access in the first few years, but after 2018 organically growing open access (non-Gold) would take over in publication counts. The top line with triangles shows the total number of publications in some kind of open access being higher (64%) than the organically growing open access in the current state (51%) by 2020.

In terms of annual cost, for comparison, the total gross nominal cost of the Research Councils’ gold (APC and subscriptions) under BIS assumptions would rise from £206 million in 2011 to £222 million in 2015 and fall to £218 million in 2020 caused by the compensating effects of higher open access and lower subscription costs. By comparison the current state of open access indicates UK spends £155m in 2011, rising to £175 million in 2015 and £196 million in 2020. The additional cost of Research Councils’ gold starts at a hypothetical £51 million above baseline in 2011 but falls year–on-year, reaching £22 million in 2020 (full estimates in Table 1).

Table 1

ECONOMIC ANALYSIS OF OPEN ACCESS OPTIONS (£M)

ALL GROSS NOMINAL CURRENT £M

	2011	2012	2013	2014	2015	2016	2017	2018	2019	2020	Cum
Current State of Open Access in UK
APC	12	15	17	21	25	30	37	44	52	63	316
Subscription	143	145	148	149	150	150	148	145	141	133	1452
Total	155	160	165	170	175	180	185	189	193	196	1768
OA/UK Coverage	11%				24%					51%
Gold Access to organically growing OA in UK
APC	32	39	47	56	67	80	96	116	139	167	839
Subscription	143	145	148	149	150	150	148	145	141	133	1452
Total	175	184	194	205	217	230	245	261	280	300	2291
OA/UK Coverage	11%				24%					51%
Gold access to all Research Councils’ publications
APC	66	68	70	72	74	76	79	81	84	86	756
Subscription	140	143	145	147	148	148	146	143	139	132	1431
Total	206	211	215	219	222	224	225	225	222	218	2187
OA/UK Coverage	35%				43%					64%

Figure 1

Projected refereed publications under constant growth. UK total and organic open access.

Figure 2

The potential impact of Research Councils’ gold option for increasing overall OA. Projected publications under constant growth: UK total publications, UK OA organic if no RC gold is implemented and UK total (organic plus RC publications) if RC gold is implemented.

Figure 3

Research Councils gold potential. Pay for gold access to RC publications line (circles), allow for organic OA to overtake (squares) to add to a higher total than organic OA alone would achieve (triangles). RC publications would all be gold, and other publications may be gold or green but OA...

Knowledge & Innovation Analysis Team;
Department for Business, Innovation and Skills;

31 January 2013