OLH gives Evidence to Wellcome Trust Review on Open Access

Posted by Martin Paul Eve on 22 May 2018

Today, the 22nd May 2018, the OLH’s Martin Paul Eve, Professor of Literature, Technology and Publishing at Birkbeck, University of London, gave evidence to the Wellcome Trust's review on its open access policy.

Eve said: "it is heartening that the Wellcome Trust, a key promoter of open access, has opened itself to a critical evaluation of its policy. We hope that the Trust will continue to ensure that work that it funds -- and academic work more broadly -- receives the widest dissemination through open access channels".

The OLH's evidence was as follows:

1. Wellcome, like many funders, supports a transition to universal OA. How is your current business model changing or how may it change to accommodate universal OA?

Unlike other publishers – such as Springer Nature, who recently flipped an open-access journal to a subscription model (“As of January 1, 2017, IMMI will convert from an Open Access journal to a subscription journal” SpringerNature, Integrating Materials and Manufacturing Innovation), warned its investors that changes to rights ownership (i.e. open licensing) may be a threat to its profits, and stated that it intends to use Impact Factor to raise its APC levels (SpringerNature, Prospectus Dated April 25, 2018) – the Open Library of Humanities only uses its business model to support a transition to open access.

The Open Library of Humanities is funded by a non-classical consortial economic model. Many institutions (just under 250) contribute a sum that is less than a single article processing charge elsewhere. This allows us to professionally publish our 23 journals in a fully gold open-access format, with open licensing (CC BY in most cases except where difficult with third-party permissions), without any article processing charges for the author.

The OLH is a charitable company limited by guarantee with objects of advancing public education through the publication of open-access research material. This means that, unlike for-profit entities, we have no conflict of interest in implementing our business model and achieving universal OA. Our model also has no conflict of interest between ensuring the quality of outputs; unlike an APC-model, the OLH model has no incentive to publish sub-par material since our revenue stream does not depend upon a per-article fee. Finally, we allow our supporting members to vote on whether or not to allow new admissions.

On this front, we have a strict policy on admissions at OLH that any new journal joining our platform must conform to the following requirements:

  1. Are peer reviewed.
  2. Have been established for at least five years.
  3. Are currently funded through a subscription model.
  4. Are based in a humanities discipline, as self-defined by authors and editors.
  5. Have an international editorial board.

Point number three is here intended to ensure that we only ask for additional funding from our library members/funders when there is the potential for offsetting. However, according to the University of California, other publishers such as Elsevier have hindered financial offsetting by only permitting substitution of titles, rather than a reduction in price, for their big deals (Fortney). This unidirectionality nonetheless helps in a move towards universal OA.

In terms of changes to our business model, we do not intend to implement any radical changes in the near future. We believe our model is already well-suited to helping with the transition to universal OA, particularly in disciplines such as the humanities subjects where Article Processing Charges do not scale well (especially when there is a localised economy [the UK] moving faster towards OA than the global one). Our model was born-OA from the start. It is scalable and ethical in its sole support for open access.

The only changes/expansions that we envisage, then, are:

  1. To continue to scale our model, including growing the library consortium so that costs are spread, and the number of titles that we publish. In doing this we provide strong evidence that OA can work in the humanities.
  2. To continue to encourage learned societies and other publishers to adopt our distributive model that spreads costs between many entities while never excluding an author on his or her ability to pay an article processing charge. For instance, the French-Canadian academic platform Érudit is exploring whether they could adopt a model similar to ours. We believe that spreading this model to multiple entities would be a sound way to normalize this model and have provided some initial thoughts for learned societies on how they could flip to our model (Eve).

2. Some funders and institutions have decided that they will no longer fund APCs (or set price caps) in hybrid journals.  Should Wellcome follow suit and why?

Wellcome should no longer fund hybrid publication but it should not relax its compliance criteria as a result. It should do so to lower its own publication costs but also to encourage a transition to universal open access, with which hybrid is not currently helping.

It is a well-known fact, acknowledged in the very announcement for this panel, that hybrid journals are more (34% on average) expensive per open-access article than pure-gold OA (Pinfield et al.; Kiley). It is also well known that some publishers, such as Elsevier, continue to make the argument that there is no “double dipping” in their model, since, in a phrase attributed to Alicia Wise in 2014, “money coming in through a journal subscription is used to pay for a particular number of articles, and that open-access articles in hybrid journals are additional to that” (Smith). The official policy of this publisher is that “Adjustments in individual journal subscription list prices will be based only on criteria applied to subscription articles”, that is, explicitly, that there will be no reduction in subscription costs or subscription volume based on the quantity of hybrid OA material published (Elsevier, Pricing).

In other words, at present, the hybrid model appears to be seen by many publishers as an additional revenue stream, with no desire for offsetting. Why, when Elsevier claim, as above, to perceive hybrid as simply an additional source of revenue, would they make changes to their model that transition to universal OA, as per question 1, above? Indeed, this publisher does not even appear to have an offsetting agreement in place with Jisc in the UK (Earney).

While hybrid encourages researchers to publish in open access (by ensuring that researchers do not have to change any of their current publication practices/venue selections), it has been ineffective as a transition strategy. This is because hybrid only positively contributes to the reputation of journals that receive high-quality Wellcome-funded research. As Springer Nature recently put it: “Some of our journals are among the open access journals with the highest impact factor, providing us with the ability to charge higher APCs for these journals than for journals with average impact factors” (SpringerNature, Prospectus Dated April 25, 2018 59). Elsevier likewise notes in its official pricing policy at the time of writing that “Impact Factor” is one of the core considerations for the level of APC that is charged (Elsevier, Pricing). In other words, publishers use Wellcome’s hybrid-funded work to raise the prestige of their titles. They then use this symbolic capital/prestige to raise their OA fees, payable by the Trust.

Indeed, the Trust has a great deal of power in its research funding on which it should capitalize in order to keep its publication expenditure at a sensible level, while also encouraging a transition to universal OA. As an aside in a personal rather than institutional capacity, I can state (as Professor of Literature, Technology and Publishing at Birkbeck) that in my experience grant capture is a much larger and more tangible factor in promotion and hiring than the venue of publication outputs. It is not likely that top researchers will be dissuaded from taking Wellcome money and conducting research if hybrid venues are ruled non-compliant. Who pays the piper calls the tune.

It has also been argued that hybrid is essential to academic freedom and that a funder determining publication venue is somehow in violation of this. The history of academic freedom, though, does not match up with the present situation. Take, for example, the UK’s Education Reform Act of 1988, in which academic freedom is legally enshrined “to ensure that academic staff have freedom within the law to question and test received wisdom, and to put forward new ideas and controversial or unpopular opinions, without placing themselves in jeopardy of losing their jobs or privileges they may have at their institutions” (Education Reform Act 1988). In one reading, of course, this could be seen as endorsing the free choice of where to publish one's research. In another take, though, one that situates academic freedom within a history of censorship, this statement instead refers to the ability to publish the work without fear of institutional or government reprisal, but not to choose where to publish it. The freedom to publish the results. Given that academics already play so many games to satisfy their institutions and funders with regard to publishing in supposedly desirable venues, if one reads this in the former light, “freedom” is surely already compromised through willing submission to playing such games. If one reads it in the mode of the latter, then the complaint is that a mechanism designed to avoid academics being prohibited from disseminating their research is compromised by asking academics to ensure the broadest dissemination of their material by OA. This simply does not make sense.

As a born-OA pure-gold publisher with an innovative business model, we wish to see the subscription ecosystem dry up and to move towards a world in which all research is available to read, so that it may do the greatest good in the world. We believe that hybrid is an expensive strategy that does not and cannot, by the explicit admissions of other major publishers, fulfil this goal.

3. Many researchers and institutions argue that there is too much friction in the system, making it difficult to understand and comply with publisher and funder OA policies and processes.  Do you agree and if so, how could Wellcome seek to reduce this through its OA policy?  

It is extraordinary that researchers who work upon some of the most complex problems in the contemporary natural sciences should find it difficult to understand and to comply with the contract of their research funder, particularly when many such scientists have academic institutions behind them to support this activity.

At least part of the problem of complexity, though, is the unwillingness of publishers to change their policies to match the Trust’s requirements. For instance, Elsevier requires an entirely separate page on its website simply to outline how to achieve compliance (Elsevier, Wellcome Trust (UK)). One could argue, then, as this question presumes, that it is Wellcome’s fault for having a policy that is different to other norms. On the other hand, had Elsevier’s journals policies that were simply compliant by default, this would also achieve a frictionless mechanism. For instance, the publisher guidelines here state that “When publishing under either one of these open access options, the corresponding author will be given a choice of Creative Commons licenses”, yet “Wellcome Trust policy states that authors are required to select a Creative Commons Attribution (CC-BY) license” (Elsevier, Wellcome Trust (UK)). A way of “reducing friction” here would be for Elsevier: 1. to license all its material in a way that is compliant, removing the need to know this from authors; or 2. to remove this “choice” for Wellcome-funded researchers.

We would further add that it is not a bad thing to introduce some element of friction into scholarly workflows. As above in our response to question 2, hybrid is a way of reducing friction, but it has not led to the transition aimed for in the objectives of this review. The fact that researchers may have to learn something about the venues in which they choose to publish is not necessarily a bad thing.

4. If you were tasked to rewrite Wellcome’s OA policy – and cognisant of the objectives we have defined – what one thing (if any) would you change and why?

In addition to eradicating hybrid funding, we would recommend that the Trust commit to a hard line on open access – and using its financial clout to ensure that its funded researchers turn only to pure-gold venues, which in turn encourages a transition – but that it provide a variety of funding mechanisms to achieve this. Specifically, we would recommend removing the contingent clause “wherever such options exist for their publisher of choice and are compliant with our policy” as this is a clear get-out in some cases. We would also recommend the rephrasing of “open access charges” to a styling that supports innovation in business models to lower the costs of scholarly communications.

Works Cited

 Earney, Liam. “Offsetting and Its Discontents: Challenges and Opportunities of Open Access Offsetting Agreements.” Insights the UKSG Journal, vol. 30, no. 1, Mar. 2017, pp. 11–24. CrossRef, doi:10.1629/uksg.345.

Education Reform Act 1988. http://www.legislation.gov.uk/ukpga/1988/40/part/IV/crossheading/academic-tenure. Accessed 9 July 2014.

Elsevier. Pricing. 2018, https://www.elsevier.com/about/our-business/policies/pricing#Dipping.

---. Wellcome Trust (UK). 2018, https://www.elsevier.com/about/open-science/open-access/agreements/elsevier-agreement-with-the-wellcome-trust.

Eve, Martin Paul. “How Learned Societies Could Flip to Open Access, With No Author-Facing Charges, Using a Consortial Model.” Martin Paul Eve, 21 Jan. 2018, https://www.martineve.com/2018/01/21/how-learned-societies-could-flip-to-oa-using-a-consortial-model/.

Fortney, Katie. “UC Linguistics Faculty Pledge Support for Glossa, Call for Cancellation of Lingua.” Office of Scholarly Communication, 16 Feb. 2016, http://osc.universityofcalifornia.edu/2016/02/uc-lingustics-faculty-support-glossa/.

Kiley, Robert. Wellcome Is Going to Review Its Open Access Policy. 2018, https://wellcome.ac.uk/news/wellcome-going-review-its-open-access-policy.

Pinfield, Stephen, et al. “The ‘Total Cost of Publication’ in a Hybrid Open-Access Environment: Institutional Approaches to Funding Journal Article-Processing Charges in Combination with Subscriptions.” Journal of the Association for Information Science and Technology, vol. 67, no. 7, July 2016, pp. 1751–66. CrossRef, doi:10.1002/asi.23446.

Smith, Adam. “Interview with Alicia Wise.” Research Fortnight, Nov. 2014, p. 6.

SpringerNature. Prospectus Dated April 25, 2018. 2018, http://proxy.dbagproject.de/mediacenter/ressourcen/pdf/emissionen/springernature_prospectus.pdf.

---. Integrating Materials and Manufacturing Innovation. 2018, https://link.springer.com/journal/40192.