January 23, 2017

Study: Journal Bundle Discounts Vary Widely Across Institutions

Most academic librarians are familiar with the ‘big deal’ bundles offered by large academic publishers, which grant access to a large number of journals from a particular publisher at a discounted rate. And many will also be familiar with the opacity surrounding those deals, which are often negotiated on a school-by-school basis with confidentiality clauses in place. A new study of the economics of these bundle deals suggests that variations in how these bundles are priced for different institutions mean that they are a better deal for some schools than others.

Led by University of California Santa Barbara economics professor Theodore Bergstrom and published in the Proceedings of the National Academy of Sciences (PNAS), the study found that bundle prices, which are not generally made public, “differ drastically between institutions.” To learn more about the typically confidential pricing plans, the research team first had to file Freedom of Information Act (FOIA) requests to numerous publicly funded universities throughout the United States to get the contracts the schools had signed for bundled journal deals with six publishing houses—Wiley, Sage, Taylor & Francis, Springer, Elsevier, and Emerald—and three nonprofit publishers—the American Chemical Society, Cambridge University Press, and Oxford University Press.

The result of those requests was more than 360 contracts for bundled journal subscriptions from 55 university libraries and a dozen library consortia. The universities were broken down into three camps according to their 2000 Carnegie classification as Research 1 universities, which grant more than 50 PhDs annually; Research 2 Universities which grant between 20 and 50 PhDs; and Master’s schools, which granted more than 50 Master’s degrees and fewer than 20 PhDs every year.

To accurately judge how good a deal these journal subscribers were getting on these bundles, researchers used the ‘cost per citation’ of a given bundle of journals to measure its cost effectiveness, to address the fact that highly cited journals are regularly bundled alongside others that are rarely used. As an example of these big deal bundles, the study cites Elsevier’s “Freedom Collection,” which offers access to 2,200 titles from the publisher. Tellingly, the bottom fifty percent of the collection is responsible for just five percent of the citations it generates.

Cost per citation is also key to buyers in judging the real value of the discounts given by bundle deals, said Ivy Anderson, director of collection management and development for the California Digital Library (CDL), which manages online subscriptions for the University of California system. While the breadth of the programs the UC system offers means that the organization has a reason to subscribe to even niche journals, even CDL has found itself forced to cancel bundle packages that weren’t seeing enough use to validate their cost. “On one level, every bundled license is a great deal–lots of content for a relatively discounted price,” Anderson told LJ. “But the reality is that high use does tend to cluster in a very small number of journals, and then there is a very long tail.”

The study found that publishers charge different institutions differently for access to their journals, with large institutions that grant many PhDs paying more for access, on average, than their smaller, less research-driven counterparts. According to some commercial publishers, those differences are similar to the tiered pricing practiced by some nonprofit publishers. “We have aimed to be consistent about what individual universities pay extra for the content they were not previously purchasing,” a spokesperson for Sage told LJ. “This doesn’t mean all institutions pay the same amount, as we have applied tiered prices to take into account differences such as [full-time enrollment] and research intensity.”

In addition, the PNAS study found that even at discounted prices, commercial journals didn’t provide the same level of value to research institutions that nonprofits did. “The value for money that universities are getting from commercial publishers is generally considerably less than the value for money they are getting from nonprofit publishers,” study author and University of Michigan economics professor Paul Courant told The Guardian. LJ contributor Dorothea Salo also discussed the implications of the study in a recent column.

The study also points out that the reluctance of some librarians to drive hard bargains with publishers may help to keep prices high, a problem that is exacerbated by the secrecy that often surround bundle pricing. The study cited CDL in particular as on example where “bargaining has led to significant price reductions for big deal bundles.” That may be one reason that between 2006 and 2012, the number of Association of Research Libraries (ARL) member libraries subscribing to full-list bundle packages from commercial publishers has steadily declined. Those cancellations are good for no one, leaving libraries without subscription access to research while publishers get hit in the bottom line.

While those cancellations are painful, though, they could serve as an important wake up call to publishers. According to Ivy Anderson, “we may just have to experience the pain of reduced access as a way to confront the affordability problem. That may in fact be our best hope of introducing true market forces into a very distorted industry.“

Ian Chant About Ian Chant

Ian Chant is a former editor at LJ and a freelance journalist whose work has appeared in Scientific American and Popular Mechanics and on NPR.