According to Ross Atkinson, associate university librarian for collections at Cornell University, the decisions on cancellations have been made on the basis of faculty input, as well as several years of statistical information on individual journal use. He said that the library's budget simply couldn't handle the cost increase of Elsevier's bundled journals package, and therefore Cornell was moving to a title-by-title plan. While the result will mean the library will pay more per title, Atkinson said it would gain some savings. The library and the university faculty have been preparing for the cancellations for over a year.

The full situation that Cornell has been facing with its journals is explained in a section of the library's Web site called "Issues in Scholarly Communication" (http://www.library.cornell.edu/scholarlycomm/problem.html). The statistics for the top North American research libraries over the last 15 years, provided from the Association of Research Libraries and posted on the Cornell site, are rather telling: "[T]he prices of serials have increased by 215 percent, library expenditures on serials have gone up by 210 percent, and the serials titles purchased by large academic research libraries have decreased by 5 percent. The Consumer Price Index during the same period has increased by only 62 percent."

Basically, Cornell has increased its budget for serials over the years but is purchasing fewer titles, and "therefore providing access to an increasingly smaller portion of the scholarly information that is being published." Cornell spends about $1.7 million a year just for Elsevier titles, which "account for less than 2 percent of the serials to which the Cornell Library subscribes, but that cost is equal to over 20 percent of the Library's total serials expenditures including the Medical School."

The situation is similar at Harvard University. According to Ivy Anderson, the digital acquisitions program librarian, the library plans to cancel a substantial number of Elsevier journals, partly to reduce print duplication across the University in response to budget constraints, but also "to regain control over collection decisions."

Elsevier's licensing model has presented a problem for Harvard. Here's how Anderson described the situation: "Under Elsevier's current pricing, the cost of online access steeply increases in response to any cancellations at all; thus we will have to cancel a significant number of titles in order to achieve any cost reductions. By changing our own buying practices, we also hope that we can influence the dynamics of the way that online journals are being sold. The large packages lock in a very high fixed expenditure that continues to rise more quickly than the rate of inflation, eroding our ability to purchase other materials. The larger the package, the more stress this places on our budgets and purchasing power—and Elsevier is the largest and most expensive of these packages. In times of fiscal constraint, such as we're currently experiencing, this is becoming increasingly insupportable. It is leading to ever more homogenous library collections developed by publishers, not librarians. We need the flexibility to shape our collections in response to changing programs and budgets."

The University of California is still negotiating for the full database package from Elsevier. According to university librarian and executive director of the California Digital Library, Daniel Greenstein, there are really only two options: either Elsevier and UC work out agreeable and affordable terms or UC will be forced to trim titles.

Price increases and tough budget decisions are of course not a new phenomenon. But, Greenstein said, what is new is the broadening of concern beyond librarians and involvement by faculty leadership in the economics of scholarly publishing. Faculty members are both information producers and users and will be a key driver of change in the scholarly communications model. "We may be at a turning point," he said.

The Academic Senate at UC Santa Cruz recently passed a resolution demanding reasonable rates and urging faculty "to seriously consider cutting ties with Elsevier by refusing to sit on its boards, referee its journals, and submit papers to the publisher." UCSC currently spends half of its budget for online journals on its subscription to ScienceDirect Online from Elsevier.

Many faculty scholars at numerous universities have already embraced alternative scholarly publishing and open-access models, such as BioMed Central, Public Library of Science (PLoS), and others. Just this week, SPARC (the Scholarly Publishing and Academic Resources Coalition) announced a partnership with PLoS "to broaden support for open-access publishing among researchers, funding agencies, societies, libraries, and academic institutions through cooperative educational and advocacy activities."

I contacted Elsevier to discuss the pricing and bundling issues. According to Elsevier's director for corporate relations, Eric Merkel-Sobotta, the company sent a letter to customers in June stating that the average price increase would be 6.5 percent, which reflected "both inflationary increases as well as volume growth." He explained: "We don't force anyone to buy anything they don't want," but it's "clearly advantageous to take bundled offerings," just like buying a 12-pack of a soft drink rather than buying individual cans. The company is moving toward database pricing and away from print-based pricing models.

He said Elsevier is working closely with Cornell on a customized list of titles. "The vast majority of cancellations are duplicates and we are working with them so the cancellation of unique titles is kept at an absolute minimum." He also indicated that Elsevier was "close to coming to a mutually acceptable agreement with California." He added: "The atmosphere in the negotiation rooms is really much better than what you might have heard."

After talking with Merkel-Sobotta, I noticed this update posted to the Cornell site on Nov. 14:
"We have received a call from Elsevier, suggesting that we discuss further the price increases for 2004. We will, of course, be very glad to do this, since it could mean we might cancel somewhat fewer titles. If this turns out to be the case, the list of canceled titles on this site will be revised, to indicate the titles we will be able to retain for next year. These discussions will not alter the fact, however, that we are moving from the bundled to the title-by-title plan, and that the number of journals we expect to cancel will still be substantial."

These current developments are indicative of the unsettled state of scholarly communications and of the ongoing transition from print models to digital publishing and pricing models. As Greenstein stated: "How do we value a title? There is no consensus on what acceptable usage-based pricing would look like."

Note: For background information, read: "The Scholars Rebellion Against Scholarly Publishing Practices: Varmus, Vitek, and Venting," by Myer Kutz, in the January 2002 issue of Searcher (http://www.infotoday.com/searcher/jan02/kutz.htm).