The Association of American University Presses has issued a statement on Open Access. Peter Suber and Ben Vershbow have commented the statement. Also, there was an article about it in Inside Higher Ed ("University Presses Take Their Stand").
Here is an excerpt from that statement (minus footnote numbers):
From the founding of the first American university presses in the late 19th century, the purpose of the university press has always been to assist the university in fulfilling its noble mission "to advance knowledge, and to diffuse it not merely among those who can attend the daily lectures—but far and wide," in the famous words of President Daniel Coit Gilman of the Johns Hopkins University. Universities acknowledged then that for most scholarly works there was insufficient commercial demand to sustain a publishing operation on sales alone, and recognized an obligation to establish and subsidize their own presses in order to serve the mission of universities to share the knowledge they generate.
Knowledge is expensive to produce, and requires—in addition to the scholar’s own work—knowledgeable editorial selection and careful vetting as well as a high level of quality in copyediting, design, production, marketing, and distribution in order to achieve the excellence for which American universities have come to be widely praised. Universities have made substantial investments in their presses, and the staffs who run them are expert at what they do. The system of scholarly communication that these presses support has played a vital role in the spread of knowledge worldwide. Calls for changing this system need to take careful account of the costs of doing so, not just for individual presses but for their parent universities, and for the scholarly societies that also contribute in major ways to the current system.
And, indeed, while proud of their achievements, university presses and scholarly societies have never been averse to change. Rather, being embedded in the culture of higher education that values experimentation and advances in knowledge, presses have themselves been open to new ways of facilitating scholarly communication and have been active participants in the process. Prominent examples from the last decade include Project MUSE, the History E-Book Project, the History Cooperative, California’s AnthroSource and eScholarship Editions, Cambridge Companions Online, Chicago’s online edition of The Founders’ Constitution, Columbia’s International Affairs Online (CIAO) and Gutenberg-e, The New Georgia Encyclopedia, MIT CogNet, Oxford Scholarship Online and Oxford’s recent experiments with open access journals, Virginia’s Rotunda, and Michigan’s new press and library collaboration digitalculturebooks.
The phrase "open access" has come to symbolize the pressure for change in the system, largely in response to the financial burden on academic libraries of maintaining subscriptions to commercially published journals in science, technology, and medicine (STM). Without reform to this system many fear that the results of new research will increasingly be accessible to an ever-shrinking number of the wealthiest universities. Hence the call has arisen for a new publishing model of open access that will ensure the continued ability of universities to disseminate knowledge "far and wide."
The well-known Budapest Open Access Initiative (BOAI), in promoting a solution to the high price of STM journals, defines open access as "permitting any users to read, download, copy, distribute, print, search, or link to the full texts of these articles, crawl them for indexing, pass them as data to software, or use them for any other lawful purpose, without financial, legal, or technical barriers other than those inseparable from gaining access to the Internet itself." In principle, this definition of open access could be applied to all types of scholarly publishing, and calls for widespread use of institutional repositories and for self-archiving by individual scholars in order to promote such open access are by no means limited to just STM journal literature. Although the debate over open access has centered almost exclusively on one sector of publishing, STM journals, there is no reason to limit the discussion to that sector and indeed, given the interconnectedness of knowledge, it is unwise not to explore the implications of open access for all fields of knowledge lest an unfortunate new "digital divide" should arise between fields and between different types of publishing. The recently proposed legislation known as the Federal Research Public Access Act of 2006 (FRPAA) would affect a wide range of research that receives funding from 11 federal agencies, including the National Science Foundation, the Environmental Protection Agency, and the Departments of Energy, Education, and Defense. The American Council of Learned Societies, in its 2006 report on "Cyberinfrastructure for the Humanities and Social Sciences," has advocated such open access for all social science and humanities scholarship. However, there is a wide range of models that can be subsumed under the generic term "open access," with both risks and benefits to the entire system of scholarly communications that are as yet not fully understood.
The member presses of AAUP, which include scholarly societies, recognize that they have an obligation to confront the many challenges—economic, legal, and technological—to the existing system and to participate with all willing partners, both within and outside the university, to strengthen and expand scholarly communications. Many of them, often in collaboration with research libraries, are already experimenting with new approaches, including varieties of open access that seek to balance the mission of scholarly communication with its costs.
Those costs today are covered by a combination of institutional subsidies and sales in the marketplace. On average, AAUP university-based members receive about 10% of their revenue as subsidies from their parent institution, 85% from sales, and 5% from other sources. Therefore the AAUP believes it is important to keep an open mind about what constitutes open access, since some kinds of open access are compatible with a market-based model. The National Academies Press, for instance, makes all of its books available online for free full-text browsing worldwide while offering both downloadable PDFs and print copies for sale.
For the more radical approaches that abandon the market as a viable basis for the recovery of costs in scholarly publishing and instead try to implement a model that has come to be known as the "gift economy" or the "subsidy economy," the AAUP urges that the following points be kept in mind:
1) BOAI-type open access will require large contributions from either the authors or other sources (including foundations and libraries, which pay "member" fees instead of paying for subscriptions). Scholars at less wealthy institutions or those with no institutional affiliations may experience greater difficulty in publishing unless fees are waived or reduced (a process that will increase the burden on other authors, who will have to pay higher fees to offset the waivers). This will be especially true for monographs, the publishing cost for which now runs around $25,000 to $30,000 (for an average monograph of 250 pages with no illustrations) and would still be close to $20,000 to $25,000 if no printing were done or inventory maintained by the publisher. While inequities among users may be resolved by open-access publishing, they may resurface as inequities among authors.
2) Costs for scholarly communication overall will not change radically, but merely be shifted from one sector of the university to another. For university presses and scholarly societies currently, only 17% to 20% of the publishing costs of monographs are spent on manufacturing, so most of their other expenses will still need to be covered Even so, many end users will prefer to print out what they want to read, especially longer articles and books, using printing devices that are less economical than dedicated printing presses. Moreover, since traditional print publishing will not disappear overnight, there will be the continuing costs of maintaining that part of the system in addition to the new costs of supporting online publishing ventures. Finally, if faculty are asked themselves to become publishers, they will spend more of their time performing tasks for which they are not trained and less on the teaching and research for which they are, resulting in an overall loss in economic efficiency for the university as a whole.
3) Requirements for fully free-to-user open access publishing of journal articles, whether through the journals themselves or by way of open institutional repositories or authors’ self-archiving, will undermine existing well-regarded services like Project MUSE (the electronic database of more than 300 journals in the humanities and social sciences jointly operated by the library and press at Johns Hopkins) that rely on institutional site licensing to be sustained. BOAI-style open access is inherently incompatible with site licensing as a model for journal publishing and archiving.
4) In 2005, university presses recovered 90% of their operating costs, roughly $500 million, from sales. Of that $500 million, sales to libraries account for 15% to 20%, or $75 to $100 million. The rest comes from sales to general and college bookstores, to online retailers, and directly to individual scholars. Under free-to-user open access, universities that operate presses would need to be prepared to decide how much of the cost of maintaining the system they would want to continue bearing and how much they would expect other universities to absorb by providing full or partial faculty subsidies for publication of both journal articles and monographs. Any university opting for full support could expect its costs to rise dramatically. Conversely, if any parent university decided to maintain only its current support, other universities not now supporting the system, or doing so only through small and occasional subsidies for faculty publication, would also see their costs increase. Offsetting these costs would be whatever amounts their libraries would save in journal subscriptions and monograph purchases, but since commercial publishers (and many society publishers) would not have the option of converting to a full "subsidy economy," those amounts would be equivalent to only what libraries currently spend on university press publications.
5) If commercial publishers should decide to stop publishing research under the constrained circumstances envisioned by advocates of free-to-user open access, what happens to the journals abandoned by these publishers? How many of them could universities afford to subsidize through faculty grants? How much could universities with presses increase the output of their presses to accommodate the monographs now published commercially? The answers to these questions could involve significant new capital investments. In addition, the case of scholarly societies under BOAI-style open access is particularly worrying. As non-profit organizations committed to supporting effective scholarly communications and professional standards in their fields, these societies provide a wide range of services to scholars and scholarship, including annual conferences, professional development opportunities, recognition of scholarly excellence, and statistical information on such matters as enrollment and employment in their fields, as well as respected publishing programs. Whether a given society’s publishing activities underwrite other services or must be supported by other revenues, funding for essential professional and scholarly activities would be jeopardized by a mandated shift to free-to-user open access, increasing the financial burdens on individual scholars as both authors and professionals.
For university presses, unlike commercial and society publishers, open access does not necessarily pose a threat to their operation and their pursuit of the mission to "advance knowledge, and to diffuse it. . . far and wide." Presses can exist in a gift economy for at least the most scholarly of their publishing functions if costs are internally reallocated (from library purchases to faculty grants and press subsidies). But presses have increasingly been required by their parent universities to operate in the market economy, and the concern that presses have for the erosion of copyright protection directly reflects this pressure. Any decision to switch from a market to a gift economy requires very careful thought and planning. The AAUP and its member presses welcome the opportunity to collaborate with university administrators, librarians, and faculty in designing new publishing models, mindful that it is important to protect what is most valuable about the existing system, which has served the scholarly community and the general public so well for over a century, while undertaking reforms to make the system work better for everyone in the future.