Abstract
Faced with a $500,000 budget shortfall in 2012, Mississippi State University library undertook a review of two of the largest of their “Big Deal” subscription packages. Jones and Marshall describe how they reviewed usage statistics collected over the previous four years and evaluated cost-per-download. The Administration's decision to cancel the two packages and subscribe individually to approximately two hundred of the cancelled titles is detailed, as well as the impact on the faculty, students, and library services. Additionally, they caution other libraries in similar situations to take other metrics and user needs into account and to include others outside the library in the decision-making process.
The Serials and the Electronic Resources departments of the library at Mississippi State University (MSU) were informed by the Dean of Libraries in late 2011 that the budget needed to be reduced by $500,000 for the 2012 fiscal year. The Library had, in the previous four years, reduced individual serial title purchases and had cancelled six databases in the 2011 fiscal year. They had not added any new titles to the collection since 2005 with the exception of those added in a few packages and databases purchased. With the need to reduce the budget by such a significant amount, the only materials left to consider cancelling were the big package deals of serial titles. The Dean of Libraries had hoped to negotiate a smaller budget reduction and since the decision came very late in the calendar year, there was very little time for this project to proceed.
There were two packages up for renewal in 2012—Wiley-Blackwell and Springer. MSU does have additional serial packages with Elsevier (the ScienceDirect Freedom Collection), Sage, and Emerald, but those contracts were not up for renewal at that time. Mississippi State had entered into the Wiley Package in 2002, with an original purchase price of approximately $200,000. It was a shared purchase with seven other EPSCoR (Experimental Program to Stimulate Competitive Research) Science Information Group (ESIG) libraries. The Blackwell package was initiated in 2007 with an original purchase price of approximately $165,000. It too was an ESIG purchase, and it was for the whole collection of Blackwell titles. With the Wiley and Blackwell merger, MSU elected to remain in both packages. In 2010 they paid for each package separately, but in 2011 the invoice was combined at a total of approximately $400,000. The Springer package was initiated in 2007 with an initial purchase price of approximately $350,000. It was a shared title list with thirty-one other ESIG libraries.
Since time was the major factor, the team decided that they would only assess usage statistics to determine what titles would be kept or cancelled. Fortunately, they had done usage studies in 2010 in anticipation of needing to cancel titles in 2011. Library staff collected statistics from 2008, 2009 and January through June of 2010. They then ascertained the 2011 price for each title, by adding 6% to the 2010 price. Their analysis per vendor was broken down by the number of downloads per title: fifty or more, one hundred or more, two hundred or more, and three hundred or more. They then derived the cost of each title at the download level. However, at the end of the analysis for the 2011 fiscal year, the Library Administrative Council decided not to pursue any “Big Deal” cancellations at that time.
Conversely, for 2012 that was not going to be the case; cuts to the “Big Deals” were definitely going to occur. The team completed the 2010 information and added 2011 usage statistics. They knew from their previous work that they would need to look at titles downloaded fifty or one hundred times or more to prepare the data for cancellations.
The data gathered for the 2012 assessment consisted of the 2008–2010 known usage statistics, plus 2011 estimated statistics; 2011 statistics were computed by combining the known January–June data and doubling them. They calculated the price for the 2012 subscriptions by adding the contractual 5% to the 2011 price.
The results were put into an Excel spreadsheet and reviewed to determine savings for each scenario, retaining journals with fifty or more downloads per title and one hundred or more downloads per title. The team also considered multiple years for calculating numbers of downloads per title. The economics of the review showed that for Springer, there were 75 titles downloaded more than one hundred times and 190 titles downloaded more than fifty times. In both cases, the package cost more than purchasing the individual titles. For Wiley-Blackwell, there were 120 titles downloaded more than one hundred times and 230 titles downloaded more than fifty times. In this case, subscribing to the individual titles downloaded more than fifty times would cost $60,000 more than the package; however, subscribing to titles downloaded more than one hundred times would save approximately $180,000. One caveat that was noted in gathering the usage statistics is that it is important to look for title changes and publisher changes, especially when looking across years of data.
The team presented their data to the Library Administrative Council, which determined that cancelling the agreements with both Springer and Wiley-Blackwell would be the best course of action. The library then purchased individual subscriptions to approximately two hundred of the cancelled titles from the packages, based on usage data compiled over the four years. The Library saved approximately $400,000, which was as close as it could get to the target number, given the short amount of time it had to complete the project.
The impact of the loss of these two large packages has been significant. Current access to over 2,800 journals was lost. Many disciplines, especially in the Social Sciences, lost all of their titles from Springer and Wiley-Blackwell. Some small departments on campus that have fewer students and faculty than large departments have been most affected. The library did retain perpetual access to some paid subscriptions and consortia titles, but only for older materials. Old perpetual access rights do not help with access to the current material that is typically the material in high demand by the students and faculty.
If the team had to do it all over again, they would use more than just the usage statistics to determine the final mix of titles for the library. They would involve the teaching faculty, reference librarians, and liaison librarians in the decision making. As it was, the bulk of the project took place in the fall—after the date that cancellation information is due to serial vendors. The team felt they did not have the luxury of time to consult with others because of the short deadlines imposed for the project. The team would consider substantially more factors in any future cancellation project.
The feedback from the cancellation of the two large journal packages has been mixed. Many librarians wanted more information about the process and were disheartened by the loss of so much content and its impact on the library users. The reaction of the teaching faculty has been critical. They were notified via an e-mail from the Library Administrative Council and a list of cancelled titles was posted on the library website as well. Currently, librarians are meeting with each department to provide a list of current, relevant titles to the work of the department, including the costs and current usage. Departments may then elect to swap titles, keeping the total expenditure equivalent. Departmental faculty members have until July 31, 2012 to submit lists of these swap-out titles; as of early June 2012 no swap lists were received. Librarians are also providing departments with lists of ScienceDirect and Sage titles by subject, as well as promoting the services of the Interlibrary Loan department, which advertises a twenty-four-hour turn-around time to obtain material. Some of the faculty, as well as the librarians, are also beginning to be concerned about accreditation of academic departments. Librarians have offered their assistance in this process, but are having difficulty getting necessary information from the departments. Interestingly, the students have not expressed any opinion on the journal cuts.
In conclusion, the project was successful in saving 80% of the required budget shortfall. Previous years’ statistics were valuable in presenting a complete picture of the use of journals over a longer period of time, even though both packages under discussion were ultimately cancelled. The process needs more time devoted to it in order to have more involvement from librarians outside of technical services and from university faculty. Lastly, the process will likely be repeated when other big packages come up for renewal.