A Mountain of Challenges: Budgeting for Electronic Resources

10/1/2001
NELA Annual Conference
Burlington, VT

Peggy Johnson (University of Minnesota)
Carol Fleishauer (MIT)
Kathryn Poliseno (NELINET)

Presented by the New England Technical Services Librarians (NETSL) at the New England Library Association's Annual Conference, Burlington, Vermont.

Co-sponsored by the North American Serials Interest Group (NASIG).

Few libraries have the luxury of supplemental funding for electronic resources, and so must make reallocation decisions to manage constrained budgets. Budget cycles, timing, aggregations, pricing models and cost recovery, archiving and statistics were among the factors examined in this panel discussion, co-sponsored by NETSL and NASIG (the North American Serials Interest Group).

Peggy Johnson, Assistant University Librarian at the University of Minnesota, opened the panel, speaking on "Allocating and Budgeting for E-Resources." She gave five reasons why this topic should be considered important. First, electronic resources are requiring increasingly large portions of library materials budgets. In 1999, ARL libraries spent nearly 11% of their total budgets on networked resources, and that figure has now most likely risen to 25%. Second, there are multiple appended costs and activities associated with providing electronic resources, including leases and subscriptions, hardware and software, furniture, file storage and file refreshing, initial wiring and telecommunications charges, staff and user training, continuing technical support, documentation, intellectual access, interface design and redesign. This complex of factors multiplies the consequences of imprudent budget allocations. Third, since these resources are not regarded as capital expenditures, there are different kinds of accounting associated with them. A fourth factor is the complexity of costing models currently in play, including prices based on numbers of either FTE or simultaneous users, free-with-print, and others. Finally, since many administrators are still convinced that electronic resources will save money, we need to be prepared with budget information that answers this mistaken expectation. (In a humorous aside, Ms. Johnson referred to budgeting as a "Sisyphean labor," and particularly to the theory of the absurd as propounded by Camus, in The Myth of Sisyphus: "conscious man facing an unintelligible universe.")

Ms. Johnson discussed typical, traditional approaches to managing the acquisitions budget. Most libraries allocate portions of materials budgets according to the organization of selection responsibilities. These commonly may be "user groups-based" - for example, children, young adults, new readers, etc. in public libraries - or the discipline- or subject-based allocations generally found in academic libraries. Another means of allocation is the creation of a central fund for materials according to format (e.g. serials or microforms), or for materials which serve all user groups. Electronic resources are usually managed using this model. Most libraries began budget allocations with central fund for all electronic resources, particularly since the earlier resources available, such as CD-ROM databases, were general-purpose in nature. There are problems, however, with treating e-resources as though they are unique, and to which traditional policies don't apply.

Current trends in allocating funds for e-resources, by contrast, indicate a move toward distributing funds according to disciplines or user groups, paralleling such distribution for more traditional formats. At the University of Minnesota, allocations are handled in a three-tiered system. Some materials are purchased through a central fund. These are multi-disciplinary and usually large-ticket reference items. The second tier consists of items bought with two or more separate funds, when selectors agree to cooperate in the purchase of items that meet the needs of more than one subject area or user group. The third tier encompasses all items purchased out of separate budgets or fund lines. The goal is to assign personal responsibility for as many e-resources as possible. Not everything fits neatly into one of those tiers. For example, aggregators' packages and JSTOR contain titles with a narrow focus, but invoices cannot be unbundled to permit purchase through separate fund lines. A large central fund, the same tapped for multi-disciplinary e-resources, is used for this type of purchase.

Ms. Johnson addressed the subjects of allocating new funds in a fair manner, and the need for external accountability. Allocations should reflect institutional objectives and programmatic priorities, and should be consistent with policies which establish responsibility. One responsible approach to take is that new funds for electronic resources are allocated proportionately to existing funds for materials in all formats. Exceptions may be made, of course, if an area is judged to require extra funding for any legitimate reason. Accountability is involved because no matter what model you use to allocate funds, you need to be able to report total expenditures for e-resources, in terms of a percentage of the budget and/or actual dollars spent. A more difficult aspect of accountability involves how the resources are used, and what the outcomes of use are. This involves assessment, not only reporting, and addresses the issue of whether or not a resource is "on target" for one's user community.

In conclusion, Ms. Johnson stated that she finds distributed responsibility in budgeting for electronic resources preferable to the model in which all such materials are paid for from a central fund. This allows for greater accountability in budgeting, since individual selectors are responsible for materials in all formats and for decisions about allocation between formats.

A brief bibliography accompanied her handout. Ms. Johnson stated that, as she could not discover very much publication activity in this area, there are plenty of research and writing opportunities available.

Carol Fleishauer, Associate Director of Collections Services, MIT Libraries, followed with her talk, "A central electronic resources fund: progress and pressures." Twenty per cent of MIT's collections budget is dedicated to electronic products. Under an agreement with the previous Provost (and honored by the current Provost), $125,000 per year of new base-budget funds was made available for electronic resources from 1997 to 2002. In return, incremental funds for serials inflation were capped at eight per cent per year, and no other new money for collections has been made available. In essence, the agreement gave up any possibility of new book money, in order to support an aggressive posture regarding electronic resources. The allocation of funds for electronic resources, however, contrasts with those for print. In place of a main library, MIT has five divisional libraries, reporting to a single director. Each library's funds for print resources are allocated by subject in each library; electronic resources for the entire library system, by contrast, are concentrated in a single fund.

A group of collections development and reference librarians, drawn from all five libraries and wryly named NERD (Networked Electronic Resources Decision Group), helps guide decisions about purchasing from this central fund. This process applies particularly to resources which will be used by more than one library's constituency, or which are too expensive to buy with a single library's funds. (Selectors at individual libraries may spend "print" funds on electronic resources, going outside the central pool.) Criteria which are considered by NERD for recommending purchases include critical value to the whole MIT community; institutional commitment to a resource (where there exists a written or moral commitment to support a resource provider, such as SPARC); availability of full text; and a favorable cost/benefit ratio or access model.

Ms. Fleishauer spoke about the varying relationships between using funds for print vs. electronic versions of journals. Selectors may "steal" from print funds when electronic databases replace print databases. Some print journals have been cancelled in favor of their electronic replacements, but the Libraries are not quick to remove full text in print, since faculty as well as librarians are concerned about archiving. In addition, the savings that might be realized by cancelling all print subscriptions turn out not to be significant. In instances where a publisher's price base is for the electronic version of a journal, with print available as an "add-on" or at a discount, monies for the journal are moved into the central electronic resources fund.

Working with a specially-designated, central fund represents both "progress" and "problems." Problems include a loss of autonomy for individual collection decisions, and a more subjective match between purchasing decisions and program needs than obtains with formula-driven allocations. Timing of offers and deadlines from vendors, which may also be a problem with print, can be exacerbated in this model, and further complicated where consortial buying is involved. There is also the risk of diffuse responsibility for products: NERD tries to assign a librarian as "sponsor" to every product, but the sense of ownership may be reduced from what it would be when selectors spend their "own funds." It is difficult to evaluate the contributions of staff, or reward staff contributions, in this model. There can also be strains on interpersonal relations stemming from uncertainty as to whether there is really "something for everyone." On the progress side, the NERD process has resulted in excellent selection decisions and is efficient, since the group meets only once a month for two hours. The resources selected are heavily used, as may be seen from provider data, and the Web-based front end is very popular. Accounting is simplified through the flexible use of funds. There is breadth of input into selection decisions, and the overall process is educational for those involved, promoting "systems thinking" -- although for some, the latter may be a difficult requirement.

Kathryn Poliseno, Coordinator for Consortium Purchases for NELINET, rounded out the panel with some observations about the role of a regional network in helping to broker electronic resources for its member libraries. There are many interacting factors to take into account which influence what resources can be offered, and when. Prominent among these are varying budget cycles, and budgets which may be either flatlined or increased in small increments, frequently less than the 5-8% annual cost increases posted by electronic resources. The timing of new offerings is affected by varying start dates: July 1 is the most common date for libraries to begin with a new resource, with January 1 the second most common. NELINET makes an attempt to provide quarterly offerings, and tries to negotiate the most convenient start dates for the greatest number of libraries, but there is really no "good time" to make an offer.

Pricing of aggregations presents a chicken-and-egg problem. Vendors may be reluctant to discuss the contents of an aggregation offered at a given price without knowing what the level of interest is on the part of customers, but libraries are often not willing to express interest without knowing what the price is. As has often been noted, pricing models vary, and the same models may be defined differently, further complicating the process of making a good offering to potential customers. The most easily explained model is based on a projected FTE number of users. Site-licensing is also a good basis for price, but vendors may be either reluctant to define sites, or else use different criteria to define them. Other models in use are simultaneous user pricing, selling searches in blocks, and costing an electronic product as a proportion of the analogous print price.

Concerns about archival access to electronic products are still very much alive. What will a library receive if it cancels an electronic subscription? At present, one common answer is "nothing." Some offers of archival access via CD-ROM or tape are made, but it is not at all clear that these will be useful for an extended period. If an electronic archive is available from the publisher, how far back does it go, and what happens if the company goes out of business?

In short, making electronic products available to member libraries is a complicated and delicate matter for regional networks. Despite the best preparations and research, it is at present not possible to know for certain in advance if an offer will be received well by library customers.

David Miller
NETSL Writer/Editor
Levin Library, Curry College
Milton, Mass. 02186