The new Open Access (OA) policy introduced this year by Research Councils UK (RCUK) — in response to last year’s Finch Report — has been very controversial, particularly its exhortation to researchers to “prefer” Gold over Green Open Access
When it was first announced there was an outcry from UK universities over the cost implications of the new policy. In response, on 7th September last year the UK Minister for Universities and Science David Willetts made an additional £10 million available to 30 research intensive universities to help pay OA transition costs.
But the controversy has continued regardless, and in January this year the House of Lords Science & Technology Committee launched an inquiry into the policy. The subsequent report roundly criticised RCUK for the way it had been implemented, and concluded that lack of clarity about the policy and the guidance offered was ‘unacceptable’. RCUK responded by making a number of “clarifications”, and extended the permissible embargo period before research papers could be made available under Green OA from 6 and 12 months, to 24 months — an extension that led many OA advocates to complain that a bad policy had been made worse.
In the meantime, the House of Commons Business, Innovation and Skills (BIS) Select Committee had announced its own inquiry, which at the time of writing remains ongoing. During this inquiry a number of new issues have emerged, including complaints that some publishers are exploiting RCUK’s new policy to pump up their profits (profits that many believe are already unacceptably high). There are concerns, for instance, that the £10m in additional funding that Willetts provided is being used inappropriately. At the centre of these new concerns is Elsevier, the world’s largest scholarly publisher.
When last September Willetts made an additional £10 million available to research intensive universities it was widely assumed that the money had been provided to help them meet the costs arising from the fact that when the new RCUK policy came into effect on April 1st this year their researchers would have to start paying to publish their papers.
This assumption was understandable: When BIS announced the grant it said the money was, “to kick-start the process of developing policies and setting up funds to meet the costs of article processing charges (APCs).”
In the same press release Willetts was quoted saying, “This extra £10 million investment will help some of our universities move across to the open access model. This will usher in a new era of academic discovery and keep the UK at the forefront of research to drive innovation and growth.”
Critics argue, however, that at least some of this money is being used to pay for papers that have already been published in subscription journals. Specifically, they cite the fact that on December 20th last year Elsevier approached JISC Collections — the organisation that procures digital content on behalf of UK research institutions — and offered, in effect, to sell back to UK universities the papers that their researchers had published with it during 2012. That is, it offered to make papers that had been published in subscription journals OA retrospectively.
Meant to help with the transition costs
Elsevier’s offer came to light on February 11th in a submission made to the BIS Inquiry by OA advocacy group SPARC Europe. SPARC complained that the £10m was being used in ways that Willetts had not envisaged when he made the money available, since Elsevier’s offer had nothing to do with Open Access. Rather it was being used to buy “retrospective back-catalogue access.” Moreover, SPARC added, Elsevier had “already been paid, in the form of subscriptions, for handling these articles”.
Others are similarly concerned. In an email, executive director of Research Libraries UK (RLUK) David Prosser told me that while Elsevier was perfectly within its rights to make such an offer, that fact was that “the research community worldwide has already paid for these articles to be published through subscription charges.”
He added, “Any cost, even if at a discount from Elsevier's regular APC charge, would basically represent pure profit for the publisher.”
More importantly, Prosser said, using Willetts’ money in this way will not address the problem it was intended to solve — that is, to help research intensive universities migrate from an environment in which their researchers are able to publish their articles for free, to one in which they will be required to pay to publish. “The £10 million was meant to help with the transition costs that the Finch Report had identified as we move towards Open Access,” he told me. “Making previously published papers open access in this way will not help put in place structures for long-term OA.”
Are these criticisms justified?
First, is it right to assume that the institutions who have taken up Elsevier’s offer are funding it from the £10m? When I put this question to RCUK a spokeswoman replied, “I do not have any information as to whether the additional funding has been used in the way that you suggest.”
In response to the same question, JISC Collections told me, “How they funded participation in the offer I don’t know – some may of course have used the extra funding from RCUK, but if so, I am sure that they would have checked first if this was appropriate within the terms of the grant. Clearly if it was not appropriate, our members would have used other funds for their participation in the offer. Indeed, there was interest in the offer from institutions who were not recipients of the RCUK funding you mention.”
Elsevier’s director of universal access, Alicia Wise evidently does believe the institutions who took up the offer funded it with the Willetts’ money. However, she says, SPARC Europe has exaggerated things, “The scale of what we are doing is quite different to the way the SPARC Europe comment to BIS suggests,” she told me. “Our retrospective agreement is in fact a small portion of the £10m in funding that UK universities received.”
She added that she was puzzled why SPARC Europe was making an issue of the offer. “We were being responsive and helpful to universities who were trying to set up and accelerate their open-access programmes in alignment with the new UK open-access policy. So it was part of a broader managed change to Open Access that flows from the finch report. We are just being constructive partners.”
For instance, she said, “In parallel to the retrospective deal we are offering research intensive UK universities the opportunity to set up a discounted prepayment scheme in exchange for a fixed volume of OA articles.”
The second thing to consider in judging SPARC’s complaint is on what basis the grant was provided. “We do not know the details of the grant conditions attached to the BIS £10m, but anecdotally we believe that these were extremely flexible,” says Wise. “Apparently SPARC Europe think that the grant was made more narrowly, but I wonder if they have seen the grant letter.”
When I asked RCUK if it could specify the conditions attached to the grant, I was told, “The purpose of the funding was to pump prime open access in universities and research organisations, and help with establishing open access policies and funds that will support academics. It was also to help provide funding for articles being published this year or in making previously published articles retrospectively open access as per the RCUK open access policy.”
Nevertheless, in light of claims that these articles had already been paid for by means of journal subscriptions, and that many of them would in any case become freely available once Elsevier’s green embargo expired (The embargo for many of Elsevier’s journals is 12 months. Moreover, authors can make their manuscripts freely available immediately after acceptance), I asked RCUK if it did not feel that taxpayers’ money could have been better spent.
“We provided the funds to HEIs in a way that was designed to be flexible and could be used to best support their transition to open access”, the RCUK spokeswoman replied. “They will be reporting to us on how they spent the funds and, if they have paid for retrospective access, they will need to explain why this, as well as other elements of how they have spent the money, was good use of funds.”
Who, why, how much?
When I asked Wise how many universities had taken up Elsevier’s offer she said that 12 institutions had responded, and that “under 700” papers had been purchased.
Wise added that the papers had been sold at up to a 50% discount on Elsevier’s standard 2012 article-processing charges (Elsevier’s APCs range from $500-$5,000 per paper). “This reflected the element of bulk buying, the number of universities that were involved, and the fact that we had had up to one years’ worth of subscription revenue for them. The fee was heavily discounted to reflect this.”
In other words, Elsevier took a number of factors into account when setting a price, and it did so in line with Elsevier’s new double dipping policy. The latter states, “Elsevier amends its journal list prices to account for each and every sponsored open access article. Our subscription customers are not charged for open access articles – we do not double dip.”
And for this reason, Wise explained, the offer had closed at the end of April. “We needed the numbers to be finalised in order to accurately reflect them in the 2014 list prices”.
For all that, it is difficult to judge whether the price that participating universities have paid for the articles is fair and reasonable, not least because Wise declined to say how much money had changed hands, or what the average price per paper had been.
Nevertheless, we could make a guestimate. If, for instance, we assume that the 700 papers had been published in journals that charge Elsevier’s highest APC level of $5,000, and if we assume that all of these articles were sold at a 50% discount, then the total would presumably have been around $1,750,000 (£1,151,543) — some 11.5% of the sum that Willetts provided in additional funding.
If, on the other hand, we assume the papers were charged at the industry standard for Hybrid OA of $3,000, then the total would be more like $1,050,000 (£690,925).
The most intriguing question, however, is why 12 UK institutions would have spent taxpayers’ money buying articles that their own researchers had produced, that the research community had already subsidised through journal subscriptions, and which would have become freely available to all in the near future?
Alas, it is not possible to answer this question. JISC Collections emailed all 12 universities on my behalf inviting them to speak to me, and giving them my details. Not one of them contacted me. Why? Alma Swan, director of European advocacy at SPARC, suspects that this is because Elsevier would have insisted that, as a condition of the sale, participating institutions would have had to sign a non-disclosure agreement (NDA) — a point to which we will return.
When I asked Wise why she thought the universities agreed to buy the papers she answered, “The institutions haven’t sponsored all of their articles for 2012. They went through quite carefully and selected which articles were of interest to them — so it might have been a paper in a particular subject area that they wanted, or it might have been in subject areas where our embargoes are longer than normal. I don’t know the basis for all their decisions, but they did select specific articles. It was not a blanket agreement.”
She added, “We then checked for duplicates to ensure that more than one institution was not offering to fund the same article where there was multiple authorship.”
JISC Collections suggested another explanation for why the universities had taken part: “Some of the institutions I spoke with considered participation in the offer because they wanted to make strategically important articles immediately open access on a CC-BY licence”.
So does this mean that all 700 papers are now freely available under a CC-BY licence, or shortly will be, I asked Wise? “We are still working through that,” she replied. “The aspiration is CC-BY, but we are still figuring out what we can actually do from an administrative perspective, how clearly they can be labelled.”
The wish to have papers made available on a CC-BY basis would doubtless have been an important factor for Elsevier when deciding what to charge. As Wise points out, “If the papers are made available CC-BY there are implications for any secondary revenue streams we would have expected to earn — from reprints, commercial reuse etc. These revenues vary from title to title but under CC-BY they would no longer be available to us.”
In addition, it turns out that freeing papers from the paywalls of Elsevier’s ScienceDirect platform is devilishly difficult. “There is quite an astonishing complexity our end in converting the articles from the original basis on which they were published to an OA basis,” Wise said. “Amongst the many things we are having to do is to figure out how to go through every article and change the licence labels associated with it. We are having to do this manually, and we have numerous back office systems. We are also having to push the articles over to Europe PMC in some cases, and in other discipline areas we are adding links to those articles. We are still figuring out internally what the workflows will be to do this.”
When I put these points to Swan, however, she was sceptical. “As far as I can see, the issue of needing to charge such a high amount for retrospective ‘OA’ is still not explained. And since most of the articles would have already been out of, or shortly to have been out of, embargoes, this is, I’m afraid, double-dipping at its worst.”
Wise clearly disagrees. She also wonders why SPARC singled out Elsevier for criticism over the retrospective offer? “Why isn’t SPARC Europe pointing at BIS or RCUK, or at the universities who accepted our offer?” she asks. We are just being responsive to requests that we receive and trying to put forward constructive solutions.”
Swan responds, “We are pointing at Elsevier because it was Elsevier’s initiative. To get its hands on as much of the Willets’ money as possible. I don’t blame it for that — as I’ve always said, these publishers are only doing their jobs. And I’m doing mine — pointing out where greed is being exhibited in return for very little advance in OA.”
Whatever the rights and wrongs of its retrospective deal, Elsevier turns out to be the only publisher to have sold papers back to the research community in this way. Indeed, other publishers declined to do so even when invited to. “We try to ensure that we are even handed with our NESLi2 publishers,” explains JISC Collections. “When we consulted with the institutions about the offer from Elsevier for making 2012 articles open access we found there was some interest in this concept. Thus, we contacted the other publishers to ask if they were interested in making a similar offer. None so far has done so.”
Be that as it may, we are still left struggling to judge the fairness of Elsevier’s offer. Not only do we not know how much it charged to free the papers, but we have no idea what impact liberating them would have had on Elsevier’s subscription revenues, or on any potential secondary revenues. “It’s all a bit of fog isn’t it?” says Swan.
Consequently, Swan makes a suggestion, “If SPARC Europe’s description of Elsevier’s offer was inaccurate, and if the process is really so complex, why doesn’t Elsevier tell the story publicly. Lots of people would be interested to hear how it’s done and to see the breakdown of costs between those tasks. Maybe the costs to Elsevier even exceed the revenue, which would be cause for having praise heaped upon it. Let’s see the cost breakdown so we can judge, and then offer up due praise if it seems Elsevier is doing all this more or less as a public service. And SPARC Europe would lead that praise.”
Swan’s suggestion draws our attention to an issue that has long bedevilled scholarly publishing — a serious lack of transparency. Several decades ago scholarly publishers all but ceased selling individual subscriptions to journals. Instead, they began bundling hundreds of journals together into large packages — a sales technique that came to be known as the Big Deal. These packages are then sold on a multi-year basis, and on the condition that librarians sign NDAs to prevent anyone else knowing what they paid.
Such lack of transparency is an issue that was raised on a number of occasions during the evidence sessions of the BIS Inquiry. And during the final evidence session Labour MP Ann McKechin pointed out to David Willetts that the non-transparency clauses that publishers use to prevent libraries from understanding the “real price in the market” of journals means that they are unable to drive down costs.
Willetts replied that while such clauses could “leave a nasty taste in the mouth” the government had no legal power to prevent their use.
Clearly, therefore, when considering OA pricing a key issue is whether it will prove more transparent than the subscription model, and so ensure a competitive market emerges. OA advocates have always argued that it will, on the grounds that the buying decisions will be made by the users of the service (authors), not intermediary librarians. In addition, they assert, these buying decisions will be made on a paper-by-paper basis.
This was a point BIS made to me when I contacted it about Elsevier’s retrospective offer. “One of the advantages in future of Gold open access is that it will make the cost of publication of research, which is an essential part of the research process, more transparent,” a spokeswoman told me. “This is because the publication funds in HEIs will be used to pay for publication in an individual journal.”
However, as SPARC Europe pointed out in its submission to BIS, the “block grant” arrangement that RCUK plans to use to fund APCs is unlikely to prove any more effective than the traditional subscription system in this regard. As SPARC put it, “[T]he proposed system of block grants to institutions will hinder the development of a more competitive market. Authors will be one step removed from the purchase decision by dint of an inevitable allocation system within each institution.”
And SPARC cited the Elsevier retrospective offer as an example of how non-transparency looks set to persist in the new OA environment. “This example clearly illustrates that the proposed block grant approach will simply turn the current ‘Big Deal’ system for subscriptions into a Big Deal system for article processing charges.”
It seems fair to conclude that Elsevier’s proposed prepayment offer will have the same effect. And the dangers implicit in such an approach became apparent recently when an Austrian researcher announced on Google Plus that he had had a new paper published. When asked if he had paid to have the paper published he replied, “They [the publisher] automatically charged FWF [the Austrian Science Fund] for it through my project. We didn't even know it was happening until afterwards … We were really surprised to notice that the article was ‘author choice’ open access after publication.”
This could hardly be described as a transparent process. And we must doubt that it is a cost effective way of doing things.
We should note that the paper in question was published not by Elsevier, but by the American Chemical Society (ACS), but (as noted) it is hard to see how Elsevier’s prepayment system would be any more likely to encourage a competitive marketplace to emerge. Nor presumably would RCUK’s block grant approach.
In fairness we should also point out that Elsevier is not the only publisher to have been accused of seeking to take advantage of the RCUK policy. During the last evidence session of the BIS Inquiry, for instance, members’ attention was drawn to an (unnamed) publisher who recently introduced a 24-month green embargo. Previously, authors were permitted to deposit their papers in their institutional repository without any embargo at all.
It was suggested to committee members that the publisher had introduced the embargo as a direct response to RCUK’s policy (which now permits embargoes of up to 24 months). As it happens, OA advocates had predicted that publishers would respond to RCUK’s policy in just his way. Nor would it have been hard to predict: scholarly publishers dislike Green OA intensely, and they dislike it because it is the only way that the excessive profits currently being made from journal publishing can be curbed. That is, Green OA would force publishers to downsize their operations and focus on the only task that is required of them in an online environment: managing peer review.
Agreeing that the sudden introduction of a 24-month embargo by a publisher that previously had no embargo was a worrying development, the BIS Committee asked Ron Egginton — who works in the Research Funding Unit of BIS — to speak to Alma Swan and investigate the matter.
But while this incident reminds us that Elsevier is not the only publisher to attract criticism, it is nevertheless the publisher most frequently criticised by the research community.
There are good reasons for this. Elsevier consistently earns profits of 36% and more (profits that in an earlier inquiry into scholarly publishing one MP suggested other companies would give their eye teeth for). And it has lobbied aggressively against Open Access — most notoriously last year when it supported the infamous US Research Works Act (RWA). If it had passed, the RWA would — amongst other things — have rolled back the NIH Public Access Policy, and so dealt a huge blow to the Open Access movement.
Unsurprisingly, it was these two reasons that were cited last year when over 13,000 researchers committed to boycott Elsevier. It was for these reasons too that on hearing that Elsevier had acquired Mendeley — the free reference manager and PDF organiser — Mendeley users immediately began to delete their accounts. And it is for these reasons that researchers keep the company under constant scrutiny.
And it was as a result of this constant scrutiny that SPARC Europe was able to bring a further issue to the attention of the BIS Committee. Referring members to a recent blog post by University of Sussex physicist Peter Coles entitled “Desperate Publishers”, SPARC suggested that Elsevier has been seeking to convince researchers funded by UK research councils that Gold OA is the default position, and that they must therefore pay to publish.
In the post SPARC referred to Coles describes how when a colleague went online to publish a conference paper via Elsevier’s electronic submission system, he was led to believe that he had no choice but to pay an article-processing fee.
As Coles explained, “When he tried to submit his article via the web upload facility he was directed to a screen asking whether his work was funded by the Science and Technology Facilities Council. When he answered “yes” he was told he was obliged to pay $3,000 for the privilege of publishing his paper in Gold Open Access mode….”
Coles added, “The publisher was deliberately and disgracefully misleading the author about the RCUK policy on open access in order to take money from them. There is no requirement for researchers to pay for Gold OA in such a case. Sharp practice is too polite a phrase to describe the actions of this publisher. And of course nobody mentioned the $3,000 fee when he signed up to give a talk at the conference.”
When I raised the matter with Wise, however, she denied it. “It is wrong to say that we are forcing authors with RCUK-funding to publish Gold OA,” she told me, adding, “Peter Coles was inaccurate in saying that when the author went in to the submission system he was asked if he was funded by one of the RCUK funders, and then told he had to pay. That is not correct at all.”
Wise continued, “At the point of submission researchers are offered the option to pay, but they can also choose the Green OA self-archiving option. We leave it to authors to choose that.”
I suggested to Wise that the best way of demonstrating that Coles was wrong would be to send me a screenshot of the submission screen. She agreed to do so and the image can be seen below.
When she emailed the image to me Wise said, “In requesting this I discovered that, in consultation with our European Library Advisory Board, we have revised the wording on our submission system to make our intentions to offer authors choices even more clear.”
I asked when the submission screen had been altered. Wise replied, “This latest revision was done yesterday [The day we had spoken]. I had no idea they were discussing this while we were doing so! There have been earlier tweaks and changes as these are live systems and we do refresh them regularly based on feedback/questions from users. Our aim is to be clear, and constructive feedback is helpful to us in achieving this aim.”
Unfairly singled out?
As I noted, Elsevier is the scholarly publisher that is most frequently criticised by researchers. When I put this to Wise she agreed, adding, “Somehow we are being portrayed as this evil force. But we are simply working in good faith to help manage the transition to OA.”
Let’s finish then by asking a question: Is Elsevier unfairly singled out for criticism?
Swan denies that SPARC has ever singled out Elsevier for criticism. It has merely drawn attention to the company’s actions and to what others are saying about it. Nor, she adds, does SPARC believe the company is evil, “There are plenty of people portraying Elsevier as evil — they are Elsevier’s customers and suppliers — but SPARC Europe is not one of them,” she told me. She added, however, that Elsevier’s frequent claims that it is interested only in helping the research community is “patent nonsense.”
I put it to Wise that if Elsevier is singled out by critics, this is not without cause. After all, the company was very antagonist towards Open Access historically, not least through its support for the RWA. Moreover, I added, although Elsevier eventually withdrew its support, the publisher has never issued a public apology for having supported the bill in the first place.
Perhaps, responded Wise, but there is nevertheless, “an unfortunate tendency for some to jump to negative conclusions without considering why we have taken a particular action.”
This is doubtless true. Moreover, it would seem that Elsevier is sometimes damned not for its actions, but for the company it keeps. A good example of this occurred on 17th May, when OA advocate and director of Harvard’s Office for Scholarly Communication Peter Suber drew attention to a misinformed attack on the White House Memorandum on Expanding Public Access to the Results of Federally Funded Research and the various state-level bills designed to expand access to publicly funded research in the US.
“I haven't seen this kind of motivated distortion since the days of PRISM and the Research Works Act,” commented Suber. “[I]t's a crude bolus of false assertions and assumptions.
Since Elsevier is a member of the organisation that published the attack — the industry group NetChoice — it was assumed that the text had been drafted by the publisher. In the event, it turned out that the executive director of NetChoice had written it.
Nevertheless, as Suber’s post began to attract the attention of OA advocates Wise, along with Elsevier’s Vice President and Head of Global Corporate Relations Tom Reller, had to move quickly to damp down a potential tweet storm protesting the company’s involvement in NetChoice.
“Just because we're a member doesn't mean we c everything,” tweeted Reller.
“[W]e did not draft/see the pub, & have serious concerns re tone and wording,” Wise added.
Subsequently, Reller told the Times Higher Education that Elsevier had expressed to NetChoice its “serious concern about the tone and content” of the posting and the “lack of transparency in the process by which [it] was developed”.
Let’s be clear: there is no reason to suggest that Elsevier is anything other than a respectable company doing what its shareholders rightfully demand of it – maximising profits. Both Swan and Prosser agree on that. But this is no longer an issue of corporate probity alone, it is also about how, in the process of pursuing its legitimate commercial interests, Elsevier increasingly finds itself at variance with the expectations and aspirations of the research community.
And once again, this is an issue that doesn’t concern Elsevier only. It also concerns the unnamed publisher that recently introduced a 24-month embargo where none previously existed, and it concerns all those other publishers who pride themselves on their aggressive pursuit of profits — companies, for instance, like Taylor & Francis (which recently saw the entire editorial board of one of its journals resign in protest), and to Wiley (currently bemoaning the fact that Science Europe has failed to adopt the RCUK model for OA, and aghast that it believes Hybrid OA as currently conceived is not viable).
As these traditional publishers begin to embrace OA they bring their aggressive commercial instincts and habits with them. The problem is that researchers have become more averse to these habits in recent years, increasingly viewing them as out of tenor with the aims and objectives of OA, and indeed of scholarly communication at large. Above all, there is a much greater awareness that scholarly publishers are enriching themselves courtesy of taxpayer’s money.
For the moment, however, the frustration for advocates of OA is that, however out of sympathy with this restless pursuit of profits researchers might now be, when it comes to getting a paper published they know that these companies still control most of the journals they feel they have to publish in if they want to advance their careers.
Amongst the papers published in the latest issue of Elsevier’s Journal of Informetrics, for instance is one co-authored by David Solomon. (The article can be purchased from Elsevier for $27.95 here). Solomon, we should note, is a founding member of the Open Access Scholarly Publishers Association (OASPA), a founder of the open-access journal Medical Education Online (MEO), and a professor in the Department of Medicine and OMERAD at Michigan State University.
Last year Solomon was one of the 13,000+ researchers that committed to boycott Elsevier. When I asked him why he had nevertheless published in an Elsevier journal, Solomon replied, “I published two articles in Elsevier journals after signing the Cost of Knowledge petition. It was a mistake to sign the petition and I feel bad about signing it and a bit of a hypocrite. I do however basically agree with the objections in the rationale for the petition.”
And that is the hugely complex and often contradictory landscape that the House of Lords Science & Technology Committee and the BIS Committee have had to struggle to get to grips with.