Policy by the Numbers

Data for sound policymaking from Google and friends

Open Access to Scholarly Literature and How to Achieve It

Wednesday, January 9, 2013

Andrew Adams is a Professor of Information Ethics in the Graduate School of Business Administration and Deputy Director of the Centre for Business Information Ethics at Meiji University in Tokyo. This is the second of two Policy By the Numbers posts by Andrew on open access publishing (see "Open Access to Scholarly Literature and Why It Matters").

The recent Finch Report, sponsored by the UK government, claimed that Green Open Access had failed to deliver substantial Open Access to scholarly/scientific journal papers and that instead extra funding should be provided to pay for Gold Open Access author fees.

Gargouri et al. put that to the test. Using data from the ROARMAP register of Mandates they clearly showed that strong mandates for deposit generate a deposit rate of over 70% of published papers within 2 years.

Many top-rated journals charge significant fees, whether the entire journal is open access or authors pay a separate charge to make their individual articles open access. For example, Springer offers hybrid open access under its "Open Choice" scheme, charging $3000 USD to make one article available freely. UK academics produce the third largest number of published articles in the world at just under 100,000 per year. According to the Finch Report, the average author charge is more than two thousand dollars, although that figure is disputed as unrepresentative by Peter Suber. Using Suber's estimate of around $900, a very rough estimate for paying OA publication fees where demanded for all articles by UK authors is $90 million USD. Unless every other country adopted the same approach, the UK would still need to continue paying their subscriptions to journals in order to access the rest of the world's output.

The Finch committee seems to many to have been more concerned with maintaining both publishers' existing profits and their potential for increasing their profits. Academic journal publishing is currently a highly profitable busines; estimates vary, but profit margins of 30% or more were reported by McGuigan and Russell in 2008, well in excess of almost any other stable industry, including other areas of publishing, as shown by the sub-5% margins reported in that article for periodical publishing in general.

Instead of pursuing the "Gold Fever" of requiring authors to choose Open Access journals or pay for individual gold access for articles, the solution is quite simple: funding bodies and research organisations can require the deposit of papers published by their grant-holders/employees into their institutional repository. The majority of publishers and publications allow this without any embargo period on the availability being set to open access. For the rest, nearly all publishers allow open access in the repository after an embargo period (typically six to 12 months, though sometimes longer). Even for those articles within an embargo period or in a published in a journal that does not ever permit open access, repository software allows readers to make an individual request for access to the article. Such access is simple for the author to provide so long as the full text has been placed in the repository: a single click on the link in the email authorises an individual mailing of the paper, a process that journals have always allowed. The requirement to deposit, called a deposit mandate, is entirely within the reasonable rights of both funding agencies and employers, being little more than a requirement to maximise the impact of their researchers' output. It’s just a little icing on the cake of requiring publication in the first place to justify funding, promotion or tenure decisions. The University of Liege and the Belgian funding body FRS-FNRS have adopted mutually reinforcing deposit mandates held up as the ideal by many open access archivangelists: researchers must deposit their author final text—not the publisher's layout, just text/illustrations as peer reviewed—immediately upon acceptance for publication; this deposit must be in the institutional repository; evaluation of researchers for promotion at the institution or future funding can refer only to deposited papers; and access to the deposits must be set to open where publisher policies allow.

High Energy Physics, a field which has had near 100% open access for well over a decade via the central repository of the ArXiv (easily compatible with institutional deposit since the software can automatically copy records between the institutional repository and ArXiv) has demonstrated that repository access does not automatically lead to the collapse of journals. No HE physics journal has gone bankrupt nor even seen substantial drops in revenue so far. There is no reason for funders and research institutions not to adopt a deposit mandate along the Liege/FRS-FNRS model and doing so, as presented in the previous post is beneficial to science, scholarship and the reputations of both institutions and the individual researchers.