The cost of subscribing to academic journals
Much has been written about the Research Works Act [you could do a lot worse than read Saurodpod Mike on the subject], academic publishing and the relationship between the scientists who do most of the work and the publishers who then assert somewhat draconian rights over those works. A boycott of the biggest publisher of them all, Elsevier, started to gain a fair degree of traction with almost 8000 scientists having pledge to limit some or all of their interactions with Elsevier and its journals.
One of the allegations levelled at Elsevier is that they charge such exorbitant prices for subscriptions to their journals that they essentially force university libraries to subscribe to so-called “bundles” or “deals” that allow access to huge swathes of titles. Accessing all those titles individually would be prohibitively costly for any institution but by offering bundles, STEM publishers are accused of exploiting the high prices of their most popular titles to foist titles onto users and librarians that have no need for them.
At the time (Elsevier has since withdrawn its support for the RWA though not the spirit of the legislation) there was little concrete information on the interweb about Elsevier’s bundling practices (lots anecdotal!). In order to inform any decision to boycott the publisher I arranged to meet UCL’s Journals Librarian. I wanted to find out about Elsevier’s bundles and our usage of bundle titles from someone who has to justify the massive outlay on journals to the powers that be in UCL. A bout of flu put paid to the original meeting, but last week I finally managed to spend an hour with UCL’s Journals Librarian and the Geography Subject Librarian.
What I discovered certainly clouds the sometimes clear-cut message that bundling is bad. The meeting was eye-opening in some respects and Elsevier and the other big publishers certainly don’t come out smelling of roses.
##TL;DR Bundling obfuscates high subscriptions prices on the most popular journals. This is due to title bundles being priced such that for large institutions at least it is far more cost-effective to have the bundle than subscribe to exorbitantly priced journals individually, ultimately propagating the high subscription price. Publishers have a near monopoly and they use (abuse?) this position to maximise their profits. But we the scientific community have allowed this situation to arise; funding decisions, academic assessment etc. all depend on publishing in “high impact” journals, so we continue to feed the monsters that are the big STEM publishers in a self-perpetuating cycle. Ultimately it is science and society that suffers. We need to starve the monsters of content and then demand seismic shift in the baseline of academic assessment.
UCL, a large UK, multi-faculty institution with 25000 students and some 8000 staff, does quite nicely out of Elsevier’s bundle; if UCL subscribed individually to the most-used journals it would be able to afford subscriptions to the top (by usage) 350 or so journals for the price it now pays for the bundle. The bundle brings with it access to about 1500 more titles. Most of the extra journals do see limited use by staff and students at UCL. Some 10% of “deal” titles had just 1–9 “uses” in 2011 with a further 27% begin used between 10–99 times in the same year. 923 titles are considered High Use (100–999) with a further 322 having more than 1000 uses in 2011. The bundle UCL subscribes to therefore provides for reasonable access to titles that UCL could not afford to subscribe to without the bundle deal. For the most part, there is little evidence that the Journal Librarian at UCL is being forced to take journals they don’t want; the figures above are complicated (another sign that publishers don’t really want us to know what is going on?) by the fact that they contain zero uses for journals not actually taken by UCL (why can’t these be excluded? Well, ask the publisher!)
On face value this looks like a pretty good deal for UCL and other similar institutions. What is not clear from the above stats is that at a superficial level the usage stats and Elsevier’s claim that bundling widens access to titles hides the exorbitant costs of some of the most popular titles in Elsevier’s stable. The reason UCL could only afford to subscribe to the top 350 odd titles if it did so individually at the stated institutional price is the very high costs of these journals. For example, Brain Research, which is in the top 10 most used journals at UCL, costs EUR21,440; that’s the price of a reasonable family car every year! If you are a small research institution you may not be able to afford an all-encompassing bundle like the one UCL subscribes to. That leaves you in the position of subscribing to a few journals individually. Bundling clearly acts to obfuscate the real situation that some very popular journals are priced in such a silly, outrageous manner by Elsevier and other large publishers. And after paying all that cash, can we do what we want with the content? No; we are still bound by silly copyright laws that restrict how much content people can download and use. We can’t data mine the publishers’ content to find relationships between entities or who knows what might be lurking in the vast realm of published papers. Lengthy and restrictive agreements are in place to stop scientists using the content that their institutions have paid for.
There are other problems with the deals Elsevier offers institutions. Firstly, you buy the bundle as is at the time of subscription. As new titles are launched by the publisher these are not covered by the bundle deal. Once the free access offered by the publisher (ostensibly provided to lure in scientists to read and publisher in the journal) in the first year or two lapses, institutions lose access unless they vary the bundle deal and add the new title (at a cost of course!) Other publishers, most notably Springer, do better here; they have a bundle and anything newly launched that would be available in the bundle if taken today is automatically part of any existing bundle deal. The same goes for back issue subscriptions to which many institutions subscribe as a means of getting rid of print copies of journals to free up much-need space in libraries. As Elsevier adds journals to its back-issue package, institutions need to pay extra to get access to these titles, despite having subscribed to the back-issue package. Springer, I was told, plays nicely here too; as they add content to their back-issue archive this content is immediately available to subscribers to the back-issue package at no extra charge.
Many libraries are stopping the practice of taking print titles in favour of electronic-only access, essentially to cut their archival and storage costs for all that so-last-century technology, paper. Now you’d think that by not taking paper copies of titles, institutions would be able to save considerably on the costs of bundle subscriptions? Not so; Elsevier prices its bundles on the basis of the print prices of its titles. Electronic publications attract 20% VAT in the UK but surely it must be cheaper to take electronic access than paper and print access? Not in the world of academic publishing.
With the rise of open access publishing, whereby publishers charge a processing, handling and publishing fee so that a paper can be freely accessed once published (Gold Open Access), you’d have thought that the price of journals and bundles would start to come down as more and more content has already been paid for by authors and their institutions? Sounds reasonable! Well, despite Elsevier saying that as the level of Gold OA increases they’ll cut the price of their journals and bundles, this is far from the case. There certainly hasn’t been any reduction in the bundle price UCL has to pay nor any rebate from Elsevier forthcoming (UCL signs up to five-year deals as part of a UK-wide procurement deal). Even more damning, Elsevier keeps its cards very close to its chest and is less than forthcoming with the numbers of Gold OA papers in its titles. Individual institutions are unable to see for themselves whether Elsevier is being true to its word that should OA publishing increase they’d cut their subscription prices. So institutions have to trust Elsevier; and we have every reason to trust them, don’t we…?
I didn’t ask and wasn’t told how much UCL pays for its Elsevier bundle [Update: it is a lot, a million pounds or so (€1.25M) according to several sources (video). Thanks to Ross Mounce (???)(https://twitter.com/#!/rmounce) for providing the links and info]; even if I had been told, I wouldn’t have been able to publisher details here as they have to stay in-house thanks to agreements with the publisher. UK institutions have been hit hard by currency fluctuations; Elsevier prices its journals and bundles in Euros, which has appreciated ~20% relative to the British Pound and so we’re paying a lot more now than before just because Elsevier prices in Euros unlike many other publishers.
Whatever it costs, it has already reached the point where some serious discussions will have to take place once the current five-year deal comes to an end [Not that UCL is considering dropping substantial access to journals — it will have to find the money somewhere, so what will get cut?]. The cost is already prohibitive in a world of reduced funding to universities, yet the publishers have institutions over a barrel. With UK students paying £9000 per year tuition fees from the next academic year, you aren’t going to be very popular if your students turn up just when you’ve decided not to pay to access Elsevier’s or any other publishers’ journals as an act of protest. You won’t be attracting the top talent to teach your students and conduct research if those scientists can’t access the latest journals and papers.
In the UK, JISC negotiates access to journals and title bundles with the major publishers on behalf of HEFCE-funded institutions. Would publishers like Elsevier sit up and take notice if these UK institutions all decide to take a stand and refuse to pay over the odds for access to journals? Possibly, but then UK institutions would be at a competitive disadvantage compared to their US and other European counterparts. In and of itself, this isn’t going to solve anything; scientists still need to publish their work in journals and many, especially those at the start of their career, can not afford to pay for Gold OA.
Whilst job applications, tenure decisions, grant submissions and all manner of other academic assessments rely upon, to some degree or other, publishing in the journals of academic publishers will be a necessary evil. The boycott of Elsevier may have made it sit up and take notice, and try to nip the issue in the bud. The sentiment, that academics are no-longer prepared to be abused by the publishing industry, need and indeed should not be restricted to Elsevier; the other big STEM publishers have similarly transgressed. The lobbying of elected officials must continue, to present the negative impact that the current publishing model has on scientific progress and the dissemination of knowledge and not just stop with the minor victory that is the withdrawal of support for the RWA.
Publishers need to be open about the costs of their open access fees and subscription prices need to be reduced to realistic levels. Unless they do this there will remain an imperative on scientists to lobby for the dismantling of the status quo via government legislation. The UK Government has signalled that it will seek to open up access to publicly funded research and data. A recent briefing note by the Parliamentary Office for Science and Technology lays out much of the current state of affairs and the problems that need to overcome in order to achieve the aim. This process will inevitably involve the STEM publishers; after all they do do the actual publishing! However they must be made to play fair.
This is where we can make a difference. The boycott is a start and should be continued and expanded to the other major publishers as appropriate; just because the RWA is effectively dead doesn’t mean the publishers have changed their spots. And it does us a disservice to restrict this to Elsevier. Any boycott will hopefully lead to the starving of publishers of content, ultimately reducing the impact of their titles and logically the prices they can charge for access to this content. On its own, a boycott is not enough. We must do all we can to change the status quo that is academic assessment. Until we break the link between funding decisions and measures of “impact” that are totally controlled by the publishers, scientists will still need to publish in the titles of the big publishers.
We need a sustainable funding model that allows research outputs to be made available openly at the point of publishing not at the point of access. Individuals and their institutions can’t effect this change alone. Here we need the influence and will of governments to broker a sustainable and fair solution to the publishing problem. One that is fair for all, not just in the interests of the publishers. Governments hold all the cards (well the majority of them) as it is they that fund the majority of the activity of science and could, if they wanted, break the publishers’ stranglehold by mandating open access and pushing through changes to the way research is funded at the institutional level. Publishers will do all they can to maintain their profitable business model. It behoves all of us to be vigilant to the misinformation that permeates the discussion and was seen to be rife in the recent RWA spat.
We may have won this particular battle over the RWA but victory is still a long way off.
