SPARC Europe’s Second Response to the Members of the BIS Select Committee

To: Members of the BIS Select Committee
Date: 10 May 2013
From: Dr Alma Swan, Director of Advocacy, SPARC Europe
Dear Committee Members,

Following the hearing on Open Access on Tuesday 16th April, I have a number of further points to make.

The cost of Green and Gold Open Access, and transitioning to an OA world
  1. I confirm that our modeling showed that, in an all-Open Access world where all research reports are openly available, Gold Open Access would be cheaper – so long as article-processing charges are kept at a reasonable level. This, of course, would require suitable market pressure on these charges (see below).
  2. Our modeling also showed that by far the cheapest way for any actor (university or nation) to get to that point is through Green Open Access. This has much lower costs than Gold OA (especially in the UK where the costs of the repository infrastructure are largely already sunk) and simply requires institutions to continue paying subscriptions for the journals they require while using their repositories to provide access to all their outputs. In this scenario, the issue of the UK producing only 6% of the world’s research but needing access to 100% of that corpus becomes less important. Annex A shows some further data.
    ‘Hybrid’ Gold Open Access
  3. RCUK’s policy has a stated preference for Gold Open Access, and permits paying for so-called ‘hybrid’ Gold OA, where authors can opt to pay an article-processing fee to make individual articles Open Access in an otherwise toll-access (subscription-based) journal. In these cases, unless the publisher has a sophisticated mechanism in place to reduce subscription costs in line with the additional, new, revenue from this ‘hybrid’ Gold stream, the UK taxpayer ends up paying the publisher twice – once for the subscription and again to make individual articles freely available (this is known as publisher ‘double dipping’ into the taxpayer’s pot).
  4. As we heard from Elsevier, that company has a system whereby reductions in subscriptions occur after a 2-year delay (presumably accounting practices make it difficult to shorten this timespan). The important point is that the reduction in subscriptions is effectively a gift to the rest of the world from the UK, since it applies across the board, across all subscribers. Meanwhile the UK has paid for the subscriptions and for the individual article-processing charges.
  5. The bulk of the money RCUK is allocating to Gold OA will go to publishers already charging for subscriptions, since RCUK’s unusual (indeed, unique) ‘preference for Gold’ provides the incentive for these publishers to offer the ‘hybrid’ Gold option. It is questionable whether this is a good use of taxpayers’ money.
    A functioning market
  6. We have seen three or four decades of above-inflation price increases from scholarly publishers. One of the opportunities that a move to Open Access brings is the chance to correct this and create a properly functioning market for scholarly information. There are a number of things wrong with the current academic publishing market, apart from the fact that a handful of large companies dominate it. In a subscription-based academic publishing world, the users of the products are not the purchasers: libraries make the purchase decisions. In an Open Access world where there is a per-article charge for publication, market pressure on prices could be brought to bear by a system where authors make the purchase decision and use money from their research grants to pay for publishing charges. This at least enables authors to decide on how best to spend their grants and makes them weigh up for themselves the value of different routes to Open Access.
  7. RCUK’s model, where block payments are handed to universities to pay for publishing costs, will not help improve the market. The danger is that institutions, or national bodies acting on their behalf in a consortium model, may find it easiest to accept the offers of large publishers to set up a ‘block payment’ system. A university, or a consortium, could simply pay an amount upfront to a publisher which would cover the cost of a certain number of papers from that institution to be made Open Access. Publishers will claim (rightly) that this can save a university some administrative overheads. There is some attractiveness to it for institutions, therefore, and publishers certainly prefer this sort of arrangement to having to work with micropayments from thousands of authors.
  8. What such a situation represents, however, is a move from a world of big-publisher ‘Big Deals’ on subscriptions to a world where we will have big-publisher ‘Big Open Access Deals’. The smaller (often learned society or researcher-led) publishers, and the ‘pure Gold’ Open Access publishers will be yet again elbowed out and the market will remain broken.
  9. The way to create better pressure on prices is to allow authors to make the purchase decision themselves by permitting grant funds to be used for publishing charges. This is the model employed by almost all research funders around the world (excepting the Wellcome Trust which pays the charges itself on application).
  10. We heard from Alicia Wise of Elsevier that the Competition Commission held an investigation into the academic publishing market in 2001 when Reed Elsevier proposed to purchase Harcourt, itself a large publisher.  My company, Key Perspectives Ltd, was contracted by the Competition Commission to help gather market data for its investigation, so we had more than a passing interest in the process and its outcome. The full report by the Competition Commission is here: http://webarchive.nationalarchives.gov.uk/+/http://www.competition-commission.org.uk/rep_pub/reports/2001/457reed.htm#full
  11. Dr Wise stated that the Competition Commission concluded that there was no public interest issue involved. This is not an accurate portrayal of the case. The truthful way of reporting what the Competition Commission at the time concluded is that there was a split, with two of the three panel members concluding there was no public interest issue and one concluding the opposite. The Commission decided to use the majority view as its official conclusion, but it is important to note that the level of disagreement with this by the third panel member was such that a note by him explaining his disagreement was included in the Commission’s official report at the end of Chapter 2 here:  http://webarchive.nationalarchives.gov.uk/+/http://www.competition-commission.org.uk/rep_pub/reports/2001/fulltext/457c2.pdf
  12. Since 2001, when this investigation took place, the industry has seen further mergers and acquisitions, including the joining of two large publishers, Wiley and Blackwell, a few years ago. Whilst the long tail of tiny publishers has almost certainly grown longer in the past decade, the market has certainly not become any less dysfunctional.
  13. The other issue is that of transparency. Big Publisher ‘Big Deals’ have long been a topic of discussion within the academic library community, and one of the reasons is that discussion of prices is generally not allowed: libraries are forced to sign non-disclosure agreements when they conclude such a deal. During the hearing last week Dr David Prosser, director of Research Libraries UK (RLUK), tweeted this:”#bisoa perhaps have a librarian on the panel? We could tell you all about non-disclosure clauses (although not an #oa  issue)”. (https://twitter.com/RLUK_David/status/324117270773256192)
    Policy leadership
  14. RCUK is still claiming leadership in policy development. This is belied by the facts. Despite the efforts of Janet Finch and David Willetts to persuade the European Commission and other funders to adopt the same ‘Gold preferred’ model as RCUK, none have yet done so. The policies announced since that of RCUK (Australian Research Council, the consortium of Irish funders and research institutions, the White House Office of Science & Technology Policy) are policies that mandate Green OA. Other policies in the pipeline are the same.
  15. RCUK has a table (created by Science Europe) showing the results of an analysis of European funder policies. This did not indicate funder preferences, merely recording whether Green or Gold OA are permissible by each funder. Thus the subtlety of funder preferences was lost, making it look as though RCUK’s policy was in line with many others. These others, however, are those that permit Gold OA as a route to Open Access but none of them express a preference for that and none top-slice money from the research budget to pay for any costs involved.
  16. In fact, no other policy is like that of RCUK in stating a preference for Gold OA. A number of funders permit grant funds to be spent on publishing costs if authors so choose, but this is very different from allocating money up-front purely for publishing fees (and allowing this money to be spent on ‘hybrid’ OA). Our own analysis of funder policies (worldwide, not just European) shows RCUK out on its own because of this stance. The SPARC Europe policy table is presented in Annex B.
  17. There is one outlier, the Austrian national funder, FWF. This funder has agreements with publishers to pay for Gold OA and the publisher automatically charges the funder – the author has no say in the matter (for example: https://plus.google.com/100061319416663890162/posts/KAJ6gUeST5Z Perhaps this can be counted as ‘preference by practice’ for Gold OA: most of the money goes on ‘hybrid’ OA while Austrian universities continue to purchase journal subscriptions.

 

 

ANNEX A: Costs of Green And Gold Open Access

 

These data are from the economic modeling exercise by Houghton & Swan (2012).

 

UK Costs £m Savings £m Net saving (cost) £m Benefit/cost ratio
OA publishing (Gold OA)Worldwide OAUnilateral OA (148)(148) 19917 52(131) 1.40.1
OA self-archiving (Green OA)Worldwide OAUnilateral OA (18)(18) 756 57(11) 4.20.4

 

Unilateral OA

  • Unilateral OA is much more expensive
  • Costs exceed benefits
  • Net cost of unilateral Green OA is much lower than that of Gold OA (£11m: £131m)

 

Worldwide OA

  • In a Green OA world, benefits exceed costs by 4x
  • In a Gold OA world, benefits exceed costs only by a factor of 1.4

 

 

 

 

 

ANNEX B: Research funder policies on Open Access

 

At 10 February 2013 the global picture was as below.

 

Overall numbers

Green (repository-based) OA required Either Green or Gold routes satisfy policy requirements Gold (journals) preferred where available
33 14 1

 

Funders permitting Gold OA article processing fees to be paid from research grant, or by a request to the funder = 20

 

Green (repository-based) Open Access required: 33 funders

Argentina Ministerio de Ciencia, Tecnología e Innovación Productiva
Australia Australian Research Council
Australia National Health and Medical Research Council
Belgium FWO (Flanders Research Office)
Canada Canadian Institutes of Health Research
Canada National Research Council
Canada International Development Research Centre
China National Science Library Chinese Academy of Sciences
Denmark Council for Independent Research, Danish National Research Fdn, Danish Council for Strategic Research, Danish National Advanced Technology Fdn, and the Council for Technology and Innovation (joint policy)
European Union European Research Council
EU Member States EUR-OCEANS Consortium on Ocean Ecosystem Analysis
France IFREMER
Ireland Science Foundation Ireland
Ireland Health Research Board
Ireland Irish Research Council
Italy Telethon
Norway Norwegian Research Council
Spain Government of the Principality of Asturias
Spain Madrid Autonomous Community
Spain General State Administration
Ukraine Parliament of Ukraine
UK Arthritis Research UK
UK British Heart Foundation
UK Cancer Research UK
UK Chief Scientist Office Scotland
UK Department of Health
UK Dunhill Medical Trust
UK Multiple Sclerosis Society
UK Wellcome Trust
USA National Institutes of Health
USA Howard Hughes Medical Institute
USA Autism Speaks
USA Institute of Education Sciences

 

 

Either Green or Gold routes satisfy policy requirements (i.e. one or the other must be provided): 14 funders

Austria FWF (Fonds zur Foerderung der wissenschaftlichen Forschung)
Canada Ontario Institute of Cancer Research
Canada Fonds de recherche du Québec
Canada Canadian Health Services Research Foundation
Canada Heart and Stroke Foundation
European Union European Commission
EU Member States CERN
Hungary Academy of Sciences
Hungary Hungarian Scientific Research Fund (OTKA)
Iceland Rannis
India Council of Scientific and Industrial Research
Sweden Swedish research Council Formas
Sweden Swedish Research Council Vetenskapradet
Switzerland Swiss National Science Foundation

 

Gold (journals) preferred where available: 1 funder

UK Research Councils UK

 

SPARC Global