Business, Innovation and Skills CommitteeWritten evidence submitted by the Political Studies Association of the UK

Executive Summary

1. Lack of consultation by Government and the Finch Working Group means that there is scant robust evidence to support the Government’s Open Access (OA) policy. This omission is particularly serious with regard to the Government’s preference for the “gold” OA model, but also leads to inadequate embargo periods for the “green” OA model.

2. “Gold” OA has been shaped by the science, technology, engineering and mathematics (STEM) agenda, without regard to the needs of other disciplines.

3. “Gold” OA proposals will cost considerably more than the claimed benefits to humanities and social science (HSS) subjects in terms of:

The likely financial costs of article processing charges (APCs).

Serious restrictions on academic freedom to publish, thereby limiting the UK’s research capacity and output.

International perceptions of UK research simply “paying to publish”.

A decline in the international standing of UK research.

4. CC-BY licencing will mean that:

Authors lose control of their work, potentially enabling it to be used in ways which could damage their reputations and/or against the express wishes of the author.

Commercial opportunities will be handed to the UK’s scientific, research and economic competitors.

Non-UK researchers will likely avoid publishing in UK-based journals because of the lack of ownership over their work.

5. There is no necessary reason that APCs charged by publishers will not rise. Universities will face financial decisions about who and what to publish, in addition to APCs being a direct subvention from the UK taxpayer and fee paying students direct to journal publishers.

6. There is little or no international evidence that the UK will benefit from “first mover” advantage particularly with “gold” OA, with many countries noting the risks this poses to their own academic and research standing and competitiveness.

7. The PSA calls on the Government to reconsider its policy and implement it differentially across disciplines, with a preference for “green” OA in HSS with an embargo period of 36 months, and with current plans piloted first in some of the STEM subjects, particularly the life sciences while further evidence is sought on the implementation of OA in HSS.

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

8. The Report of the Finch Working Group on Expanding Access to Published Research Findings, published on 18 June 2012, has been held up by the Department of Business, Innovation and Skills (BIS) and by RCUK as the definitive analysis of open access and the academic journal publishing market. Indeed, on 16 July, David Willetts, the Minister of State for Universities and Science, fully endorsed the report’s recommendations, with the sole exception of the recommendation on VAT on online publications. RCUK announced its policy on open access the following day, 17 July. Yet the Finch Working Group, which met just five times, had less than nine months to analyse an extremely complex international market involving a vast range of stakeholders, and thus was able to conduct only a very limited analysis of the costs and benefits of different OA models or to develop an understanding of how other countries might respond to British Government initiatives in this field. Since BIS and the research councils have adopted the recommendations without conducting any further market analysis, the benefits of the policy relative to the costs have still not been quantified and a proper impact assessment is lacking. It is worth noting, however, that at an AHRC-ESRC convened meeting on Open Access on 4^th February, a very senior figure in HEFCE said that he did not anticipate there to be major economic advantages to the UK of introducing open access in HSS, but simply that “Open access is right in principle” and a very senior figure in the ESRC said he agreed “that we are not just doing [it] for economic impact. This may be the driver for BIS, but not for us”.

9. Furthermore, the Finch Group did not consult widely. Dame Janet Finch attended an Academy of Social Science meeting of learned society CEOs on the 12 June, four days before publication of her report. This was the first contact our learned societies had received from the Finch Working Group and our concerns were not reflected in the report.

10. As a consequence of the failure to consult sufficiently widely, the BIS policy on OA has been shaped largely by considerations specific to the STEM community and in particular life sciences. The STEM community has been at the forefront of the OA movement and OA publishing is, therefore, considerably better established in the sciences than in HSS. The needs and concerns of the HSS subjects are in most cases quite different to STEM. Furthermore, the benefits of making HSS research findings freely available immediately on publication are considerably smaller. Had the views of HSS community been given due consideration by Finch and BIS, it is hard to imagine that they would have developed such a clear preference for the “gold” model over the “green” model.

11. The cost of processing an HSS article is considerably higher than in STEM due to both the greater average length of HSS articles (19 pages vs. 11 pages)1 and lower acceptance rates (11% vs 42%)2. Hence the level of APCs in HSS will have to be considerably higher than the figure of £1,750 used to model the costs of introducing APCs in the Finch Report, since this figure is based on the current average level of APCs, which is heavily weighted towards APCs in the STEM subjects. This means that the total proportion of research budgets that will have to be spent on APCs in HSS will be considerably higher than the 1–1.5% that the Wellcome Foundation calculates for research it funds and which is now widely quoted by RCUK. It also implies that the total cost to British universities of the transition to open access, when libraries continue to have to pay subscriptions to journals, will be higher than the £50–60 million quoted by Finch. There is of course no certainty that journals based overseas or indeed journals with substantial non-UK authors in the social sciences and humanities will make any kind of transition away from a subscription model, or even reduce prices consequent on (minority and UK-only) payment of APCs.

12. The half-life of articles in HSS (their continued utility in front-rank research) is generally much longer than in STEM. This means that the imperative to make the article freely available immediately, the primary benefit for the “gold” model cited by Finch, does not hold in HSS. Furthermore, the commercial applications of HSS research are far more limited in most HSS subjects than in most STEM subjects.

13. The benefits of “gold”, if they were to be quantified, might, therefore, be assumed to be very modest in HSS. Yet the costs are considerable. In addition to the financial costs outlined above, there are considerable costs to academic freedom and the reputation of UK research.

14. Since a great deal of research in HSS is conducted by lone researchers working on small grants that would not be able to support the cost of APCs, a large number of academics will have to look to their institutions to fund APCs. There is an obvious risk universities will prioritise the support of established “sure-bets” rather than early career colleagues, or those returning after a career break. At worst those operating outside the mainstream may find funding completely withheld. Hence the academic who does not receive funding for APCs will be forced to go the “green” route, but most of the top international journals will not offer embargo terms that comply with RCUK’s policy, and hence freedom of choice where to be published will be dramatically undermined. The “gold” model is a system that risks promoting orthodoxy over innovation and heterodoxy.

15. Quite apart from the threat to early career academics or those challenging accepted norms, the idea that university managers will become publication gatekeepers, able to determine the journals in which their staff are published and which articles are deemed worthy of the financial support necessary to be published in the journal of their choice, runs counter to the principles of academic freedom.

16. Another issue is the threat of the “gold” model to quality control. The current peer review process is robust and generally works well. The “gold” model introduces a perverse financial incentive for publishers to publish regardless of quality. The counter argument to this is that top quality journals will not want to undermine their reputation by publishing poor quality articles, but there is already evidence that in STEM subjects as long as an author can demonstrate they have met methodological and ethical criteria open access journals will publish their work. In HSS the peer review tradition is to identify work of originality and innovation rather than simply sound methodology or ethics, yet quality peer review cannot be sustained on the level of APC quoted in Finch. Either APCs will have to rise to prohibitive levels or quality of peer review will fall.

17. There are additional and serious concerns about HEFCE proposals that all outputs for the REF 2020 should be open access. Researchers will be pushed not to publish their work in the journals with the most exacting peer review standards—many of which are published outside the UK—and benchmark themselves against the best in the world, but will rather be forced to publish journals that do not uphold the highest standards of peer review, reducing the standing of British research internationally. This would be both damaging to the standing of UK research internationally and deeply undermine the academic freedom to publish in the location of the author’s choice.

18. The PSA believes that the “green” model has considerable merit and is likely to have far fewer negative consequences for academic freedom and research standards than the “gold” model. However, in setting embargo periods, much greater consideration needs to be given to the fact that the half-life of HSS articles is considerably longer than for most STEM subjects. The embargo period for HSS of 12 months as currently set by RCUK is inadequate and should be extended to at least 36 months in order to protect the intellectual property of the researcher and to continue to deliver funding to the profession through learned societies to enable it to develop young researchers.

Rights of Use and Re-Use in Relation to Open Access Research Publications

19. The PSA is deeply concerned by RCUK’s requirement that all articles published under the “gold” model must be available under CC-BY copyright licences, which allow unrestricted reuse and modification of articles, meaning authors effectively lose control of their work. The article content could be edited or abridged to such an extent that the integrity of the sources could be compromised. Similarly under CC-BY there is no protection against poor translation. Academics could find themselves, for example, in a position where their work is quoted out of context by extremist groups or reprinted in anthologies where the context might be offensive to the authors and yet, as long as the work has been attributed, they will have no legal recourse to prevent it. Yet intellectual property rights are particularly important for academics in HSS, because, unlike STEM scholars, their work rarely can rarely be protected by other means such as patents. It is also important to bear in mind that for proponents of CC-BY licences one of the primary benefits cited is the possibility of data mining but the viability of data mining HSS articles is extremely modest.

20. In reducing the intellectual property rights of academics based at UK institutions, the Government will be directly handing commercial opportunities to companies based overseas as it is highly likely that in many non-English speaking territories the content will be translated and repackaged by intermediaries and local publishers and sold for profit without the need to pay authors any royalties.

21. At the same time overseas authors, deterred by the lack of control over their intellectual property, are likely to bypass UK journals that comply with RCUK policy, compromising the international standing of UK journals.

22. The requirement for CC-BY could potentially put upward pressure on APCs as publishers will lose secondary income from commercial re-use, which they will seek to recover through higher APCs.

23. The CC-BY licence requirement should be replaced with a CC-BY-NC-ND licence, which prevents use of the work for commercial purposes and any alteration or transformation of the work.

Costs of APCs and the Implications for Research Funding and the Taxpayer

24. Clearly it is in the interests of higher education to reduce its costs, but the “gold” model does not necessarily lead to this. Transaction costs alone are going to be high and, as acknowledged by Finch, the transitional period in which universities continue to pay subscriptions for journals in their libraries (particularly those journals in which UK researchers are a small minority of authors), yet also have to contribute towards the costs of APCs, could be many years.

25. The Government has justified support for OA on the basis that “the public shouldn’t have to pay twice”, both to fund academic research and, through subscriptions, to pay for the published outputs. However the favoured “gold” model will deliver almost entirely free goods to the publishers, paid for by publicly funded APCs. This is in effect a subsidy from taxpayers and from fee-paying students to commercial organisations, since taxpayers contribute to funding academic research through the research councils and through the portion of the fees paid by students, which are used to fund the time academics commit to research and scholarship time. The Government’s argument that “the public shouldn’t have to pay twice” is belied by their own proposals: the public will always have to pay both for the research and its publication, whether the publication costs come from APCs or subscriptions.

26. Furthermore the cost to UK institutions may transpire to be well above the £50–60 million calculated by Finch because there is growing evidence, as shown in paragraphs 28–30 below, that the rest of the world will not follow the UK in supporting the “gold” model and hence competition amongst journals for APCs will not occur. Indeed, the top international journals may be able to set APCs above the current £1,750 quoted in Finch, knowing compliance with RCUK rules will require British-based researchers to pursue the “gold” route. Furthermore, as explained in paragraph 11 above, APCs in HSS journals will have to be considerably higher than the £1,750 used for modeling costs in Finch, and used by RCUK in calculating block grants for APCs, because the overheads in HSS publishing are so much higher than in STEM. At the same time, with slower take up of OA internationally, British universities will continue to have to pay journal subscriptions for far longer than anticipated by the Government.

27. All this implies that unless BIS is willing to increase funding to HEIs, funds may have to be diverted from research and teaching to publications—both APCs and library journal subscriptions.

Level of “Gold” Open Access Uptake around the World and the Ability of UK Higher Education Institutions to Remain Competitive

28. Government claims of first mover advantage for the UK arising from pushing ahead with a highly prescriptive OA policy depend on the rest of the world following the UK’s example. Yet there is little evidence that other major countries are following the lead of the UK, particularly as regards “gold”. Indeed, the debate currently taking place in the UK is alerting other countries to the risks of the “gold” model and many are rethinking their positions. It is increasingly clear that where overseas governments and funding bodies introduce an open access policy they will not prescribe or favour any particular model. This will inhibit the development of the “gold” option internationally and the “gold” route will not be available in many leading journals around the world.

29. The Australian Research Council, the main funding body for non-medical research in Australia, has opted for the “green” model; the National Institutes of Health (NIH) in the US has also opted for the “green” model with a 12 month embargo period; according to Professor Geert Bouchaert of the Public Management Institute in Belgium “Belgium (federal) and the regions have all suspended the implementation of this open access regulation until it is clear what the consequences will be”, a position also being adopted in the Netherlands, while in Germany and France funders are implementing what they see as in the national interests of their academics and publishers.

30. Many advocates of OA in the US have expressed concern that the policy being pursued in the UK and the negative response with which it is being met in wide sections of academia, is undermining support for the principle of OA. As Professor Richard Holton of MIT has said: “One of our concerns is that OA will get a bad name in the UK—and justifiably so it is pushed through in thoughtless ways—and once this happens it is going to be an uphill struggle to get people to agree to anything”.

31. As a consequence of other countries rejecting the “gold” model, the UK runs the risk of being alone in diverting funding for research to APCs. This is of considerable concern at a time when many other countries are increasing research funding and posing a greater competitive challenge to UK institutions.

32. The prospect of having to publish under CC-BY licences may even deter leading academics from taking up posts in UK institutions.

33. A final destructive consequence of the Government’s OA policy is the threat it poses to the future viability of learned societies. Most learned societies publish a range of discipline-based journals and rely heavily on this income to provide vital support to the development of the profession in their individual disciplines, thus contributing to the high international standing of British research.

Conclusion

34. The PSA urges the Government to acknowledge that significant differences exist across disciplines and that the “gold” model poses a very substantial threat to academic freedom and the reputation of British universities. It thus urges:

a reconsideration of the preference given to the “gold” model over the “green” model and to take account of the strong evidence that “green” is a more appropriate route to OA in HSS;

a review of the embargo periods envisaged under the latter model, with at least a 36-month standard;

modification of the CC-BY licence requirement under the “gold” model to a CC-BY-NC-ND licence requirement;

de-linking REF 2020 from any OA requirements;

piloting the introduction of the policy in life sciences or some of the other STEM disciplines before any decision is made about applying it to all publicly-funded research in the UK.

7 February 2013