For example, the library at Grand Valley State University in Michigan started using a PDA program from Ebook Library in 2009. That year, library patrons used 6,239 eBooks, but only 343 of them were used enough to trigger an automatic purchase. Grand Valley paid Ebook Library $69,000, but if it had purchased all the eBooks that were skimmed, it would have paid $550,000.
So, if libraries shift to a model where they only pay for titles that they actually use, how will that affect university presses, for which a significant amount of their output comes from scholarly monographs that are rarely read? Esposito estimates that approximately 25% of university press sales (which total $320 million) go to libraries, or about $80 million in sales. Roughly 40% of all library book sales in an average general research library would be eliminated with a PDA strategy in the most extreme case, which would result in $32 million in lost revenue for university presses, or 10% of their total sales.
Rick Anderson, Associate Dean for Scholarly Resources and Collections at the University of Utah, said the following at a panel at the Association of American University Presses' annual meeting in Chicago last Tuesday: “When you describe the current situation as a partnership between libraries and university presses, that makes it sound very good and noble. Here’s another way of expressing it. University presses publish books that are no freaking good to anybody, libraries buy them and put them on the shelves, where they sit and are never used by anyone, and with the money that we used to buy them, university presses publish more books that are no use to anybody. The question becomes what should be the criteria according to which we discriminate. Should it be on the basis of what our patrons demonstrably need, or should it be on the basis of what we consider to be of high quality?”
