A Letter from Sidney Verba
As of January 1, the University is eliminating a number of journals published by Elsevier. Some of these titles were
duplicate print subscriptions. Other titles were shown to have been underused over time. Harvard libraries will fulfill
requests for articles from these journals through interlibrary loan and third-party document delivery services.
The decision to eliminate these journals was the result of careful consideration over the last 15 months. It was
driven not only by current financial realities, but alsoand perhaps more importantlyby the need to reassert
control over our collections and to encourage new models for research publication at Harvard.
For more than a year, Harvard and fellow members of the Northeast Research Libraries Consortium (NERL) attempted without
success to negotiate more flexible licensing terms with Elsevier, publisher of some 1,800 journals available online
under the brand name Science Direct. Elsevier is among a handful of journal publishers whose commercial bundling
practices are squeezing library budgets. Their licensing programs require libraries to maintain large, fixed levels of
expenditure, without the ability to cancel unneeded subscriptions.
As the publisher with the largest share of the scientific (STM) journal publishing market, Elsevier journals are by
far the most expensive. Science Direct titles cost Harvard libraries $1.7 million in 2003more than four times our
expenditure level with the next largest STM journal publisher. This is nearly 6.5% of combined library acquisitions
budgets across the University.
At the same time, Harvard libraries embarked on an aggressive program to eliminate duplicate print subscriptions
across the University, driven by a desire to streamline operations, reduce costs, and foster increased cooperation among
our diverse library units. This effort has been meeting with great success.
Elsevier's 2004 contract proposal to NERL was not responsive to Harvard's objectives. Under the most advantageous terms
offered, Harvard's costs for Elsevier journals would increase by $500,000 within five years. Of greatest concern to the
Digital Acquisitions Committee and to the University Library Council was the lack of any option by which Harvard could
prune its holdings and reduce its level of spending. Libraries wishing to cancel subscriptions could do so, but only by
incurring steeply increased fees that obliterate any potential savingswhile Elsevier's revenues continued to
Toward this end, we have foregone the NERL Elsevier license in 2004 in order to regain control over Harvard library
collections in a manner that responds to the University's academic programs. Instead, the libraries will purchase online
access to Elsevier journals individually and selectively. We believe that this action is a necessary step toward the
larger and more urgent goal of restoring control over research dissemination to the research community, a goal that will
require faculty leadership.
Impact on Harvard Faculty and Students
This strategy is not without short-term impact. In order to make the transition to more selective purchasing, we will
have to economize by canceling some journals. Even if a high percentage of these cuts can be accomplished by canceling
duplicate subscriptions, this means that Harvard faculty and students will lose immediate online access to some Elsevier
journals to which we currently subscribe. Some of our peer institutions will actually have broader online access to
Elsevier titles than Harvard's. The libraries will have to fulfill requests for articles from these journals through
interlibrary loan and third-party document delivery services. If we decide to retain some lesser-used titles in print
only, patrons will have to come to the library to use them.
Although this will cause some inconvenience, experience has shown that withdrawing low-use materials is often barely
noticed. We know, for example, that more than 20 percent of our Elsevier titles are used less than two times per month,
while 10 percent are used less than once per month. Bundling has created an artificial environment that sustains
journals that might otherwise not be viable on their own. By canceling these little-used materials, funds will become
available to acquire resources that are in higher demand.
Impact on the Harvard Libraries
With short-term loss will come short-term costs. A comparison of the cost of participating in a NERL Elsevier agreement
versus going it alone shows that it will be expensive for Harvard to "buy its way out" of the deal. However, at least a
third of the necessary cancellations have been identified in the form of duplicate subscriptions, and many libraries
have indicated that they are prepared to make further reductions based on an analysis of usage data and other value
Higher fees for electronic access will have an uneven impact across the libraries. Those that are able to cancel
titles may see a reduction in overall cost, but other Harvard libraries may be forced to absorb large increases. The ULC
plans to explore ways of easing the impact of higher fees on the libraries most affected.
Although the obstacles are considerable, the strategic reasons for rejecting the fixed-bundle agreement that Elsevier
offered to NERL are still more compelling. The combined costs of Elsevier subscriptions far outrun even its closest
competitors, while prudent cancellation decisions lead only to steeper fees. Like so many other institutions, Harvard's
collections have become hostage to this situation. Declining the bundled agreement and intentionally reducing our outlay
for Elsevier titles will ultimately give us the ability to respond to the marketplace unfettered by such artificial
Harvard will achieve the freedom to exercise judgment about which publishing avenues to support in providing academic
content to our students and faculty. Reclaiming control over our collections is only the first step. We believe this
action can be a springboard for a vigorous and sustained effort to foster new models of research publication at Harvard.
This effort could take many forms, all of which will require the active involvement of Harvard's research community.
On many levels, Harvard is changing the ways in which it does business. This is as true in the libraries as it is
anywhere at the University. The changes in our digital acquisitions program will lead to questions that many of you will
need to answer. I hope that this letter is of help to each of you in that regard.
With best wishes,
Carl H. Pforzheimer University Professor and Director of the University Library