Strategic Planning For Library Collections
and Information Resources

December 8, 1995

| Problem Statement | Strategic Options | Operational Choices |

Problem Statement

The price of scholarly information has been increasing at a rate of more than 11% a year for the past fifteen years, but funding for library acquisitions among major University research libraries has averaged less than 8% per year. At the same time, there have been new demands on collections budgets, reflecting fundamental changes in scholarship: the number of scholarly books and journals published has grown dramatically; information technology has created new kinds of digital information resources, which in turn require a new infrastructure; and research advances have created entirely new academic disciplines which require library support. At Berkeley, the inability to increase the acquisitions budget since 1990 has compromised The Library's historic commitment to develop comprehensive collections. Either the University and campus must develop a budget model to address the problem of funding comprehensive library collections, or The Library must define and implement new, more realistic collection goals and strategies.

UCOP funding nearly compensated for the impact of inflation and changing academic priorities at the Berkeley Library until 1989. In the early 1970's, The University of California's Office of the President reached an accord with the campuses and the state legislature to fund University Libraries through the "Voigt-Susskind" formula. The formula determined campus library collection budgets by analyzing the information needs of the academic programs on each campus, and increased the collections budget annually to cover costs, including inflation. Until the economic downturn of 1989, Berkeley's collections budget received increases similar to those at our peers; from 1980/81 through 1988/89 the collections budget increased almost 100%, from $3,400,000 to $6,800,00. In 1990, the University ceased funding inflationary increases in the cost of library materials, even though inflation continued at double-digit rates. Since the campus has not been able to make up the difference from its own resources, Berkeley's rate of acquisition has fallen from #1 nationally in 1981 to #10 in 1994, and from #1 in journal subscriptions in 1982 to #4 in 1994. We have fallen behind our peers in both absolute and relative terms: we have many more academic programs and students than the private research universities with which we compete for faculty, graduate students and grants.

Despite the unprecedented budgetary pressures on the Berkeley campus, the dollar amount of the Library's budget for information resources (i.e., the funds to purchase books, journals, and the purchase or licensing of digital information of all kinds, e.g., databases) has not been cut. Indeed the base budget has been raised fairly regularly from campus sources, and a temporary allocation has supplemented the base for the last two years. But the base budget, even augmented with temporary funds, has not kept pace with the growing cost of information. Since 1990 the Library real purchasing power of the collections budget has declined by 26.3 percent, so that Berkeley's base information resources budget now lags behind the budgets of peer institutions.

During 1993/94, the latest year for which comparison data are available, the collections budget for the entire Berkeley campus, including Affiliated Libraries, was $9,559,943 - fifth among ARL libraries, trailing Michigan ($10,664,110); Stanford ($11,066,524); Yale ($11,460,000); and Harvard ($14,378,067). [Note: Endowments tend to protect collection budgets from the full impact of inflation at private Universities, thus The Library has made the collection endowment its top priority in the campaign.]

Given current budget priorities and projected budget resources, it seems likely that the funding gap between Berkeley and its peers will continue to widen, and the purchasing power of the budget will continue its decline. To fully compensate for inflation - that is, to stay at the same level of collecting - would require adding $1 million each year to the base budget for information resources. The campus does not have discretionary resources of this magnitude, and The Vice Chancellor's initiative to discuss restoration of funding from UCOP will require several years - if it is successful at all. The current crisis, in short, requires us to create a new collections strategy for the Library, so that limited resources are spent as effectively as possible to meet the information needs of the campus. Even if resources if this magnitude were available, The Library would have to rethink its collections policies to adapt to structural changes in scholarly communication:

  1. The growth of information worldwide means that comprehensive collections are not really possible; even a budget that does keep pace with inflation delivers an ever decreasing percentage of relevant information to campus faculty and students.

  2. Changing patterns of scholarly inquiry, including emerging disciplines and multi-disciplinary research, create new literatures and draw on existing information resources in new ways. As a consequence, collections resources must be reallocated to support emerging fields.

  3. Changing academic priorities and programs on campus, accelerated by more than 200 new faculty appointments in the wake of VERIP, require certain new kinds of scholarly information and, by extension, an end to the development of some traditional collections.

  4. Digital technologies are making possible new forms of scholarly information and publishing, and create opportunities for libraries to make information accessible whether they own it or not.
Therefore, the Library must initiate a campus wide discussion to formulate a new strategy and budget model for delivery of library information to the campus.


Our planning assumption is that the dollar amount of the current collections budget will not increase substantially during the next five years. Projecting the historic 11% annual inflation rate into the future, in five years we will be able to buy less than half of what we buy today, just as today we are buying half of what we bought in 1985. Given this stark reality, how should the library allocate its collections and operations budgets?

  1. Should the library continue to cut collections across the board, as has been the tactic during the past five years, or should excellence in a few priority areas be protected, by cutting selectively? Who will identify those priority areas? What level of library service is acceptable for non-priority academic programs?

  2. Should the library greatly expand cooperative collection development programs with other institutions, particularly Stanford and UC? What kinds of delivery services will provide acceptable access for collections located elsewhere?

  3. Can digital collections and networked access to information substitute for ownership of print collections in some fields? If so, how will a campus infrastructure to provide access digital collections be funded? What are the organizational implications of digital library collections for the relationship between the Library and IS&T? Between the campus and the UC system?

  4. Historically, library services have been subsidized, and appear free to library users. Is this financial model viable in the future? If not: Which services should be subsidized, which should be fee-for-service? Which library users should pay, which should be subsidized?

  5. In the absence of comprehensive collections and services, how might we fulfill our mission as research collection of last resort for other libraries in the state?

  6. Up to now the Library has collected broadly and deeply, without undue concern for the unequal costs of different collections, or with unequal levels of use. How should the quality and priority of library collections be measured in the future?
These are fateful questions for the Library, yet the failure to sustain comprehensive collections will be irreversible. While we must proceed carefully, we must proceed soon: to balance the budget we must begin cutting collections by March 1, 1996. These are political questions in the best sense of the word, questions about the Library as a public good which is shared by the highly specialized research and teaching programs of the entire University: How should these questions be answered?


Within a constant-dollar budget, we must learn how to balance a selective, less comprehensive array of local collections with an exceptionally high level of access to information we do not own. Clearly the ratio between ownership and access must vary considerably from one sector of the curriculum to another (e.g. the Humanities rely more on time-depth collections; certain engineering and science fields depend more upon immediate access to the most up-to-date information). The Library's budget model, for the near term at least, is likely to include elements from several of the options described below; they are presented separately only to clarify the nature of the choices which must be made.

Should cuts be allocated across the board or selectively?

As a practical matter, The Library must select a methodology to cut the collections budget in Spring 1996. The size of the cut is not entirely known at present, but must include outcomes of the normal budget process, the 11% inflation factor, the end of temporary funding for collections, and decisions about reallocations to reflect changes in campus academic programs.

Over the past five years, The Library has made incremental cuts in information resources budgets for the various disciplines. Although some suballocations have been adjusted to meet new academic program needs, most budgets have been cut across the board by the estimated rate of inflation. This strategy, based on the belief that budgetary constraints would be temporary, is no longer defensible, if it ever was, because of its serious long-range impact on all academic programs, especially when projected into the next five years. This strategy is defensible only if the campus is willing to accept diminishing library information resources to levels far lower than in the past, or is unable to identify its academic priorities.

The alternative to across-the-board cuts is to reduce support for some collections in order to maintain or increase it for others. For example, the Library could favor high-use collections and those closely linked to Berkeley's current academic programs, reducing support to archival and research collections of record which are national archival resources. Or, over the next four years the Library might try to protect thirty five collections from inflation by adding 46% to those budgets (to be identified as academic priorities in some manner), by cutting the allocations for all other collections by 44% (in addition to the impact of inflation in those fields).

Whether the cuts are across the board or selective, the Library must continue to seek ways to ameliorate the impact of inflation: seeking new revenue sources to support collections, by building endowments and new entrepreneurial ventures; refining collection strategies through close consultation with the faculty; developing closer ties to Stanford and other University of California campuses to augment our collections; and by lowering overhead costs. None of these measures will produce revenues or savings approaching the need, but all are necessary.

A Service Model for Meeting Information Needs

Taken alone, neither alternative suggested above is acceptable: over time, across the board reductions in information resources will result in institution-wide mediocrity, and any attempt to maintain inflation-adjusted budgets in some fields and not others will result in selective mediocrity. One of the underlying reasons that historically all of our programs are ranked so highly has been the overall excellence of our library resources. If there are not the fiscal resources to sustain the traditional model of comprehensive collections across the board, the Library must develop innovative new service models to meet the information needs of its faculty and students. In fact, innovative services began to grow as soon as Library collection budgets began to be cut: as acquisitions decreased 48% over the past decade, inter-library borrowing increased 107%; cooperative collection development with Stanford has developed into a close partnership between the two Libraries; and the Berkeley Library has became a leader in developing new information technology based services. But this process must intensify.

As the Library moves away from the traditional, primarily collections-based (i.e. acquisition and inventory) model of information resources it must develop a service-delivery or performance measure of quality. While in the past the quality of The Library was measured by how much we owned, how much we acquired and how much we spent, under the new model, quality would be measured by how quickly The Library responded to requests for information. Current information needs of faculty and students, as measured by indicators such as collection use and user satisfaction, will shape our allocation of budgetary resources. Inevitably, a new model will imply dramatic changes in the ways that faculty and students will use the library:

In the short run, therefore, this strategy would depend upon highly efficient cooperative collection building with other libraries and commercial document delivery services. Ultimately, the implementation of an online library strategy would require the full development of digital information resources that are still in the experimental stages in most fields: the development and acceptance of high quality electronic scholarly journals; the development of cost effective commercial document delivery services; the development of a comprehensive campus information infrastructure; and unprecedented partnerships with other research libraries. Given this context, the performance model of Library services will proceed in different fields at different rates, proceeding quickly in Engineering and Business, soon in some of the Sciences and Professional Schools, perhaps more slowly in the Humanities and some Social Sciences.

How should we choose?

We recognize that there are no easy decisions about how to allocate resources among local collections, nor about how to achieve an optimal balance between ownership of information resources and dependence upon access to information held elsewhere. It is clear, however, that the various options are not mutually exclusive. It will be necessary to institute cuts in collections while at the same time investing in an infrastructure to expand access and instituting performance measures. Over time, it is probably inevitable that the Library will charge for some services (although many current library services are subsidized and appear to be free to library users, some services are already available only for a fee: the costs of the Baker delivery service are partially recovered through fees; photocopying is self-supporting; and some libraries are partially supported by state contracts and user fees). However, since these choices represent fundamental changes in the way we have operated they should not be adopted without due deliberation. The Library proposes this three step planning process:

1. Establish a dialogue with the campus community.

Communicate the situation to the faculty; develop a common understanding of the options; describe the possible elements of a new information delivery approach; analyze costs and benefits of the options; differentiate strategies for different collection areas.

2. Define requirements for a new information access approach.

Seek grant funding to develop implement performance measures of the quality of library information resources and services; develop budget allocation models responsive to campus academic priorities; work even more closely with our natural partners (Stanford and the other major libraries in the UC system) to develop collaborative access and collection development strategies; develop a plan to use document delivery services, analyzing costs of various levels of service and developing new funding strategies; analyze differential impact of ownership or access in various academic fields; assess the feasibility of the necessary technical infrastructure.

3. Plan and implement new budget model.

Reallocate collections to balance budgets. Seek new funding models (fees, revenue, endowments, participation by Deans, etc.). Invest in infrastructure for access (by reallocation, grants and state sources). Develop a regular process for ongoing evaluation, reallocation, and budget justification for future years.

- Peter Lyman, University Librarian
December 8, 1995

Copyright © 1997 by the Library, University of California, Berkeley. All rights reserved.
Authored by Peter Lyman, University Librarian, 12/8/95
Document maintained by: Ann Moen
Last update 1/8/98. Server manager: Contact