HOW TO PAY FOR OPEN ACCESS Frederick Friend JISC Scholarly Communication Consultant Honorary Director Scholarly Communication UCL [email protected].

HOW TO PAY FOR OPEN ACCESS

Frederick Friend

JISC Scholarly Communication Consultant

Honorary Director Scholarly Communication UCL

[email protected]

DO WE NEED TO PAY FOR OPEN ACCESS?

Surely it costs nothing for researchers to self-archive their work? Cost is incurred in setting up and maintaining a repository and in

depositors’ time. We need to be able to justify that cost if questioned by university managers.

Publishers claim that repositories are parasites, living upon the investment they make in peer review and editorial control.

This is unfair criticism because the academic community is paying for those costs through library subscriptions and through unpaid referees, but if we are to answer that criticism we need to know the full costs of self-archiving and of gold OA to compare with the cost of subscription publishing - and then relate those costs to the economic value of benefits.

We can pay for open access through non-commercial business models using internal costing and accounting.

HOW MUCH DOES IT COST TO SET UP AND RUN AN INSTITUTIONAL REPOSITORY? (1)

Alma Swan’s chapter on “The business of digital repositories” in “A DRIVER’s guide to European repositories” provides a few examples of start-up and running costs for repositories in US, Canada, Ireland and UK http://www.driver-repository.eu/PublicDocs/D7.2_1.1.pdf

Huge variation in start-up costs according to whether free software used, number of staff-days required to set up system, and which functions included in costs (e.g. advocacy?)

UK LIFE Project provides considerable detail on costs for acquiring, storing and preserving various content types in repositories, with caveat that costs given are “illustrative rather than absolute” www.life.ac.uk

LIFE Project report projects costs up to ten years ahead: peaks in costs mainly due to preservation over a ten-year period (i.e. repository costs may not follow a flat trajectory)

HOW MUCH DOES IT COST TO SET UP AND RUN AN INSTITUTIONAL REPOSITORY? (2)

“Economic Implications of Alternative Scholarly Publishing Models: exploring the costs and benefits” by John Houghton and others http://www.jisc.ac.uk/publications/publications/economicpublishingmodelsfinalreport.aspx

Taking the highest figures from other sources, Houghton adopts GB£100K p.a. as an estimate for publications-oriented institutional repository costs

The Houghton Report also estimates the cost of providing access to journal articles through self-archiving rather as £14 per article

To provide overlay services such as peer review would cost an additional £1125 per article

PAYING FOR INSTITUTIONAL REPOSITORIES BY RELATING COSTS TO BENEFITS (1)

The “income” side of the repository business model may be calculated by estimating the value of the benefits the repository provides to the institution and to individuals.

Benefits are often described in very general terms, e.g. “having a repository to hold a university’s research output advertises a university’s academic strength”, “content in a repository is cited highly”.

Are these general statements sufficient to justify to a university finance office an expenditure of several tens of thousands of euros/dollars/pounds per annum?

Can these benefits be quantified in financial terms? Should they be? Some people would argue that we should not make the attempt, that repositories have to be justified solely on their academic benefits.

Even if repositories can be justified solely on academic grounds, without costing the benefits, we still need to be more specific in describing the benefits and not rely upon general claims.

PAYING FOR INSTITUTIONAL REPOSITORIES BY RELATING COSTS TO BENEFITS (2)

Academic benefits can be described in specific rather than general terms, e.g. the benefit to a single researcher of open access to a particular journal article held in a repository and not available in the researcher’s library; the benefit to a university’s recruitment of students in a particular subject through awareness of research reports in the repository.

Stories told by individual researchers about the value of repository content can be powerful arguments, i.e. the benefit may not be quantifiable as cash but it is still specific.

Some of these stories of academic benefits may be turned into financial benefits, e.g. the extra income to a university through attracting fee-paying students.

Financial benefits may consist of cost savings or additional income for the institution.

HOW MUCH DOES GOLD OPEN ACCESS COST?

Publication charges set by publishers vary widely (this is not surprising while authors are unaware of costs and of what services they receive for the OA payment)

The typical current charge appears to be around 2000 euros per article accepted for publication

Competition for authors will bring more transparency into publishers’ OA publication charges

Even with competition, some journals will be able to charge a higher OA publication charge than other journals due either to their prestige or to the level of service they provide to the author

This will complicate estimating the cost of OA publication in advance – but neither is estimating the cost of subscriptions in advance an exact science.

PAYING FOR GOLD OPEN ACCESS: THE MECHANISMS

Handling thousands of individual payments will be expensive for whoever does it

Collective payments are best for those administering the payments, although they leave the author unaware of the cost

Collective payments can be administered either by a library (e.g. the BMC subscription model) or by a university research office (e.g. a funding agency model) or by a consortium (e.g. the Springer model)

The individual payment model could still be workable for universities using Full Economic Costing but it will not be popular with publishers

Whichever model is adopted universities need to set up a fund for the payment of OA publication charges, recovering the cost in benefits later (N.B. universities already pay library subscriptions up-front so this would not create a precedent).

PAYING FOR OPEN ACCESS: THE BENEFITS WHICH MORE THAN JUSTIFY THE COST

For researchers the benefits from OA publication are- easier access to previous research publications as they begin their research- higher citations of their work once it is complete- more feedback from other researchers reading their work.

For funders and research institutions the benefits from OA publication are:- the research they fund has greater impact than has been achieved in the past- the research cycle - where work is published, read, cited and then built upon by other researchers - is enhanced and accelerated- open access makes it easier to discover previous relevant research, not only avoiding duplication in future research but also assisting future researchers to build upon past studies- innovative collaborations between publicly-funded and commercial research organisations can develop when research papers from academic organisations are openly available.- internal administrative mechanisms for reporting an institution’s research output can be made more efficient.

CALCULATING THE VALUE OF OA BENEFITS

John Houghton uses a way to calculate OA benefits which is not familiar territory for librarians: the economic benefit to R&D

Governments are used to thinking in this way – e.g. a motorway may cost a great deal of money to build but it brings measurable economic benefits

From enhanced access and efficiency Houghton calculates £172 million p.a. In increased returns to the public sector R&D, £124 million in increased returns to HE R&D, £109 million in increased returns to UK Government R&D, and around £33 million in increased returns to Research Councils’ R&D

In the first year the returns would be 1.5 times the cost of OA publishing, after first year 5 times the cost

Funding agencies could divert 3.5% of their research funding to author-side payments before the benefits were exhausted (N.B. Wellcome estimate that no more than 2% would be required to switch to OA)

HOW TO PAY FOR GOLD OA: THE MECHANISMS

The money to pay for gold OA is already in the HE budgets The money could stay with the library budget, with libraries

purchasing OA publication charges instead of subscriptions Or the money could be transferred from libraries to research offices,

or even to individual researchers (latter option may not be very practical but it would make researchers aware of cost)

Subscription big deals could be turned into gold OA publication charge big deals, with the advantages of national negotiation and bulk purchase

Or we could be in a mixed economy of subscriptions and gold OA (cf. Springer deals)

Or we could treat gold OA payments as individual institutional subscriptions negotiated locally (BMC model)

Probable that several models will be trialled as we are in new territory

THANK YOU FOR LISTENING!

You can find information about all of JISC’s work at www.jisc.ac.uk “Economic Implications of Alternative Scholarly Publishing Models:

exploring the costs and benefits” by John Houghton and others http://www.jisc.ac.uk/publications/publications/economicpublishingmodelsfinalreport.aspx

JISC Scholarly Communication Group web-page http://www.jisc.ac.uk/aboutus/committees/workinggroups/scholarlycomms.aspx

RIN Briefing Note on Payment of Publication Fees http://www.rin.ac.uk/files/payment-of-publication-fees.pdf