Last April, David Prosser, Executive Director of Research Libraries UK (RLUK), a consortium of 30 British and Irish research libraries, put his foot down in a letter to England’s Guardian newspaper. The letter referred to the libraries’ big deals, agreements under which they buy access to bundles of journals‚ some of which are core to their institutions’ missions and others that they don’t want at all‚ with the goal of saving the time and money it would take to subscribe to titles individually. RLUK maintained at the time that price increases seen as part of the agreements over the previous three years had been exorbitant, with some libraries paying more than ¬£1 million ($1.58 million) per year. RLUK demanded price reductions or else, mainly from Wiley-Blackwell and Elsevier, large companies involved in scientific, technical, and medical publishing, the greatest drivers of cost in the big deals.
Fast forward a year and the deals have been renewed‚ RLUK’s press release claims that the group saved more than ¬£20 million ($31.6 million) per year, but the exact terms are undisclosed. As the recent academic spring has shown, though, the current arrangements still aren’t popular. Scientists have long been unhappy with the publishers that disseminate most of their research results, pointing out that some operate with 35 percent profit margins while getting free raw material (papers) and labor (writing and peer reviewing) from scientists.
A case study from the UK
It was timely, then, that David Beales, engineering librarian at California State Polytechnic University at San Luis Obispo, presented an investigation of overthrowing the big deal at the Electronic Resources in Libraries conference in Austin, TX, last week. Until October of 2011, Beales was Engineering Team Leader at Imperial College London, a position in which he worked on RLUK’s negotiations with Elsevier and Wiley-Blackwell in the research libraries’ renewals of their package deals with those companies.
Beales explained that in November, 2010, the Joint Information Systems Committee (JISC), negotiating on behalf of RLUK, rejected the first offers from Elsevier and Wiley-Blackwell. Imperial College worked with the universities of Liverpool, Manchester, and Cambridge, and with Prosser, mentioned above, to figure out what Beales referred to as Plan B‚ how to maintain access to the articles the libraries needed, outside the big deal framework.
Plan B, which would have been put into action in January 2012, was a combination of individual subscriptions to titles, backed up by document delivery for individual articles requested by users that were from titles not part of the individual subscription list. In January 2011, as a first step in creating a new model, the group requested a JR5 report from the two publishers involved, which would show the number of successful full-text article requests by year and journal. Wiley-Blackwell’s report was delayed until September 2011, the deadline for negotiations to finish. Elsevier provided the necessary data right away, though, so that initially the model was based on figures obtained from that company.
Another important part of the negotiation, and part of the alternate plan, was to negotiate prices in pounds sterling (British currency). RLUK had previously been paying Elsevier in Euro and therefore the consortium’s costs varied with currency fluctuations; in 2008, when the currency crashed, the big deal cost Kings College, where Beales worked then, 30 percent more than expected.
In June, 2011, the model was unveiled to RLUK members and by that September, each of the 30 libraries had used it to create lists of the journals they would subscribe to and figured out how much of their budget remained for document delivery charges. The spreadsheet allows a library to combine pricing information with local priorities and with usage data to figure out whether title-by-title purchasing will work out better for the institution than its current subscription bundle. The model anticipated 2012 prices by multiplying 2011 prices by a factor representing one year’s inflation. Since the libraries most recent usage data was from 2010, they multiplied that by a factor representing two years’ usage increase to estimate 2012 usage. Completing the puzzle were 2012 budget figures and document delivery charges, and the libraries’ holdings data.
Looking at life without Elsevier
When Imperial examined the titles it bought through the big deal with Elsevier, it found that only 17 out of 1,750 were journals that the academic departments at the college said they couldn’t do without and that the library couldn’t afford to buy individually. Other than those 17, the library could afford all of its core titles without relying on document delivery, since most requests are for materials from a minority of journals.
Based on the libraries’ previous experience, the model estimates that patrons will request only ten percent of PDF turnaways (articles that can’t be provided as PDFs at their library) through document delivery. If the big deal were dropped, Imperial estimated that this ten percent would come to 14,575 articles per year. About ten percent of these 14,575, says Beales, are available as free, open access materials; and the British Library, which provides PDFs within 24 hours, would provide another 65 percent of the materials (hitherto it had been the main supplier but was undergoing its own negotiations with Elsevier and had to make cuts). RLUK decided to create and fund its own document delivery service, which would take three to four days per article, and would acquire the other 25 percent of the annual shortfall. The upshot was that RLUK libraries could manage without Elsevier’s big deal, but timely access would suffer.
Wiley was another story
When analyzing Wiley-Blackwell’s deal, Beales used a different alternative: cost-per-download of articles, rather than subscriptions, as the primary acquisition mechanism before relying on document delivery. This was because this big deal involved medical information, and downloads offer the more timely access that is necessary in this field. The data from Wiley-Blackwell was surprising: it showed that if Imperial College dropped its current arrangement with the company and opted for downloads, the library would need to request only about 1000 articles per year through document delivery, and it would save 15 percent of its budget, leaving the library the money to buy about 53,000 PDFs from the non-Wiley-Blackwell journals on its wishlist.
Sticking with the status quo
So, why did Imperial College (and then the other RLUK members) decide to stick with the big deal? As mentioned above, the exact terms offered by the publishers haven’t been made public, but according to Beales, the college had a few reasons other than cost for maintaining its current arrangements. Since some of the data was late, it would have meant making a very quick decision, which the college was unwilling to do. Another consideration was that Imperial found it, as Beales said, “a bit rude” to have led other RLUK institutions through the negotiations and then abandon them. The library also wanted to know what the impact would be on workloads and workflows and on access over the long term, and what the future would bring in terms of increased availability of open access materials.
In short, there is more work to be done before Beales feels that big deals can be dropped. What’s next on the list? RLUK is developing a similar, though modified, model that will be easier for libraries to use on their own. We’ll be watching.