Issues in Science and Technology Librarianship
Acquisitions & Electronic Resources
Northern Illinois University
A curious thing happened in the wake of dismantling several of our library's Big Deal packages -- nothing. Despite cutting thousands of titles, our usage data remained constant or even increased for each of the platforms involved. On the surface it would appear that the removal of the Big Deal canopy had negligible impact. However, that is not an accurate depiction of how things actually played out. To assess the significance of the rise and demise of the Big Deal requires a more subtle and nuanced narrative.
The debate about Big Deals began almost simultaneously with their creation. Ken Frazier, then library director at the University of Wisconsin, Madison, coined the term (Frazier 2001). His central point, that the trade-off between having electronic access to a publisher's full stable of titles in exchange for a guaranteed cash-flow was inequitable and ultimately debilitating for the library, was premised on two key arguments. First, a package license agreement entailed the weakening of a collection by the importation of a mass of titles that were neither needed nor wanted. Second, a multi-year license reinforced the privileged position of commercial publishers in the marketplace by imposing ironclad restrictions against cancellations even in times of budgetary stress. Frazier's critique elicited a quick rebuttal from Tom Peters, who claimed that librarians and faculty were not particularly astute as selectors and did not possess any kind of a priori intuition into local information needs (Peters 2001). Peters was less interested in the fallout from importing journals that betrayed low usage than in discovering titles outside the "core" list that attracted high usage. His argument throughout was that a licensed package would always generate more usage than individually selected titles. The Big Deal debate continues to revolve around these two poles: on the negative side it is argued that multi-year licenses are unsustainable and a deterrent to choice and innovation, while the positive position emphasizes maximizing access to resources.
Northern Illinois University's first foray into the realm of a multi-year electronic license came in 2003 with Elsevier. That license did not constitute a Big Deal in the pure sense of the term, since access hardly extended beyond a core group of subscribed titles. The advantage to Elsevier was that the firm was guaranteed a steady cash flow in an unsteady serials market, while saving the substantial cost of printing and mailing journals. The benefit for NIU was access to a discount on the quoted print price for the journal, access to an extended and valuable backfile, a capped inflation rate, and a modest flat fee on interlibrary loan transactions.
The library administration did not enter into the agreement without some doubts. Indeed, in retrospect the library may have struck a poor bargain: sharper negotiations might have yielded a broader pool of titles. However, back in 2003 it was far from clear how the transition from print to electronic serials would play out or what kind of incentives would be used to sweeten the pot. It was abundantly clear that a discount to switch from print to electronic was a windfall for the publisher. At the time, 10% was the typical rate for both commercial and non-profit presses for so-called flip pricing. Like many other institutions, Elsevier's presence at NIU placed it at the top of the heap in terms of STEM content and overall serials expenditures, making it a high priority target for bringing into the electronic fold. Moreover, Elsevier would not allow the purchase of their pre-1995 journal backfile packages without a license. There were multiple incentives to forge an agreement.
A year later, in 2004, NIU signed license agreements with Kluwer/Springer and Wiley. The terms of these contracts were more generous and very similar in that they both hinged on pooling titles among participants in IDAL (Illinois Digital Academic Library), Illinois' academic library consortium. Online access to Wiley journals jumped from 42 titles to 258. (The number of titles would more than double after the Wiley-Blackwell merger in 2006.) Once again the license featured a capped inflation rate and a secure backfile in exchange for a no-cancellation clause. In 2010 the two agreements -- now Springer and Wiley/Blackwell -- morphed into full Big Deal arrangements where Northern enjoyed access to virtually the full journal output of both publishers. The three-year contract was premised on retention of a so-called core list of titles and a capped inflation rate. Springer and Wiley/Blackwell were more than willing to offer a large portfolio of titles in return for a predictable, guaranteed revenue stream.
It is possible to reconcile the negative and positive traits of the Big Deal by recognizing that although licensed packages pose significant economic risks, they can also reap valuable rewards. Frazier's critique of the Big Deal implicitly assumed that a license has a kind of permanence. However, there have been numerous instances of institutions abandoning Big Deals. Libraries can conceive of the license package in the relatively short term (the NIU licenses were limited to three years) as a means of identifying weak core titles and high performing non-core titles. While NIU certainly did not sign on to Big Deal agreements with this thought in mind, in retrospect this is the way things played out. It's ironic to think of jettisoning a package as a positive thing, but we emerged from the experiment with a robust set of serial titles -- especially among STEM fields.
The motive for exiting licenses that excluded provisions for extensive cancellations was strictly budgetary. Maintaining the license placed too great a burden on the remaining pool of serials. By 2009, Elsevier was absorbing a disproportionate amount of the library's materials budget, a state of affairs that had been allowed to fester far too long. Contractual inflation rates and a large "content fee" in the event of dropping out forced the library to reduce our ScienceDirect expenditure by two-thirds.
Fortunately, the exercise was not as difficult as initially thought. COUNTER usage data showed that a substantial portion of ScienceDirect titles betrayed low usage. Based on a document delivery cost threshold, we developed a list of titles to drop. A number of faculty -- especially among STEM disciplines -- grumbled about the operation but there was little opposition when we could demonstrate that some journal articles were costing $100 each to download, and sometimes considerably more. We were also flexible with the threshold. The number of Elsevier titles retained was ultimately reduced by two-thirds. Remarkably, usage numbers across the remaining titles held steady and have even increased recently, despite the cut.
The negligible impact on usage of a steep serials cut is even more dramatic when viewed in the context of the Wiley/Blackwell and Springer cancellation projects. Here we encounter full-blown Big Deal packages entailing the removal of thousands of titles. At its peak the Wiley license allowed online access to current content for more than 1,600 titles. Over the course of the contract two issues became evident: a high percentage of the titles demonstrated minimal use; and a significant percentage of the "core" titles showed low or unacceptable use. We largely reconfigured the Wiley core list. From a base of more than 300 titles 170 were retained, and another 144 were allowed to lapse. In addition, 65 new titles were brought in from outside the former core. Despite the loss of well over a thousand titles, usage data remained relatively constant. Indeed, a dozen of the top 50 most-downloaded journals from 2015 were derived from the former non-core list.
The numbers for Springer are even more striking. From an original core list of 127 titles, only 45 survived the cut based on the same cost per use threshold. At the same time, about 60 titles were retained from the non-core list. As with Wiley, the loss of thousands of titles did not dramatically impact usage numbers.
In the end, NIU collections greatly benefited from the Big Deal licensed packages, which allowed the library to run a long-term empirical test of where our patron's interests actually lie. This is particularly the case when a title was formerly designated as core yet on examination turned out to be of marginal value. It was humbling to learn that core titles that were retained at great length and at great expense performed poorly. Working in tandem with faculty to identify individual titles to cut is always a tense process that is often put off. In the meantime turnover among faculty causes research agendas to evolve and change. Yet too often these subtle shifts are not accurately reflected in serials collection decisions. We should have been able to estimate pent-up demand but there was not an adequate incentive to make a switch until we had COUNTER numbers in hand. Now we have adequate -- if not indisputable -- evidence to drop underperforming core titles in favor of new subscriptions. There is no precise measure for projecting the future performance of a particular serial, but the numbers bear out that overall usage has remained steady -- and in some instances increased -- despite massive cuts into the overall pool of serials. In this sense, the signing of a Big Deal license shone a light on serial performance that we had not had before. It may even be beneficial to initiate future Big Deal packages every decade or so -- just to assess the ongoing shifts between journals we think of as core and those we presume are less important, and then repeat the entire process.
Frazier, Kenneth. 2001. "The librarians' dilemma" contemplating the costs of the 'Big Deal'". D-Lib Magazine 7(3). Available from: http://www.dlib.org/dlib/march01/frazier/03frazier.html
This work is licensed under a Creative Commons Attribution 4.0 International License.