Last week we discussed the new “cost-per-circulation” (CPC) model for public libraries — in which they can make e-books available to patrons and pay the publisher per “loan” instead of paying fixed fees to “acquire” titles as if they were print books (the “pretend it’s print” or PIP model). HarperCollins has just become the first major trade house to license its titles to libraries under the CPC model, and a growing number of library e-book platforms now support it.

This is a major shift in public library e-book distribution, and I explained last week, it’s great for library patrons… in theory. Yet as I’ve heard from several people who use CPC-supporting libraries since last week, the reality is that CPC merely replaces one set of limitations on e-book availability with another. The CPC model may end up giving publishers more control over the titles that libraries make available and reduce libraries’ traditional curatorial roles. In other words, CPC may end up being better for publishers than it is for libraries.

Libraries’ budgets limit their e-book offerings under both CPC and PIP models. Under PIP, the library can set a fixed “acquisition” budget for e-books each year and decide which titles to offer and how many “copies” of each to get. Under CPC, libraries’ e-book expenditures become unpredictable. Many CPC-supporting libraries are placing limits on the number of e-books they will “lend” per time period to avoid blowing their budgets: they cut off “lending” when they’ve reached their daily or monthly limits and resume when the next cycle starts. As a result, patrons of CPC libraries have to figure out when their libraries’ budget cycles renew and get their e-books quickly before the library turns off the tap.

Furthermore, PIP libraries can select from a wide variety of titles, including (as of 2014, at least) frontlist titles from all major publishers, albeit sometimes at prices that are several times what consumers pay. They get to curate their catalogs to fit the needs and tastes of their local audiences, just as they do with print materials. With CPC, libraries get access to all e-books that participating publishers make available, but currently that doesn’t include anything from major publishers except some HarperCollins backlist titles.

Prices that libraries pay per circulation under the CPC model are not publicly known, but they probably vary — and they will undoubtedly vary more widely as more major publishers get involved and (perhaps) start to make frontlist titles available under this model. Libraries do know something about their cost per circulation under PIP. A 2011 study, for example, put the average figure at $1.15 for e-books, almost exactly the same as for print, while older inquiries put it at less. But comparison of pricing under PIP vs. CPC is apples and oranges — and will continue to be so until the catalogs that publishers make available under the two models are similar.

Yet that day may never come — despite the fact that some in the library community have been pushing for CPC-like models for years. The esteemed former Harvard librarian Robert Darnton advocated them back in 2009 in his book The Case for Books: Past, Present, and Future, and others have been touting its advantages over PIP since then. But libraries should be careful what they wish for.

What CPC ultimately does is give more control over e-book distribution to publishers. Let’s start with the fact that libraries never actually buy e-books — they license them. That means that publishers are not obligated to make them available to libraries as they are with print books under copyright law under the first sale doctrine (known as exhaustion outside the United States).

Libraries could make a plausible argument that the PIP model makes e-book licensing so much like print book acquisition (which is why I like Eric Hellman’s term “pretend it’s print” better than the industry jargon “one copy, one user”) that it should be treated the same under law.  This would give libraries the right to acquire whichever e-book titles they want and (probably) pay consumer prices for them. Indeed, the American library community pushed for such a change last year through a U.S. Copyright Office review of the relevant provision in copyright law. But the CPC model makes it much more difficult, if not impossible, to make that argument.

Then there’s pricing. With PIP, publishers can set whatever prices they want, but those prices usually bear some relationship to consumer e-book or hardcover print prices. With CPC, there’s no implicit relationship to consumer pricing — at least in the U.S., where there is no market for trade e-book rentals, which is what library e-book “loans” really are. So publishers will find it easier to price e-books dynamically according to their age, popularity, and even perhaps the location of each library or the demographics of its service area. They’ll be under no obligation to make CPC pricing flat or even predictable.

This will make it even harder for libraries to budget for e-book distribution. It may even lead libraries to suppress availability of certain highly popular e-book titles, because they will be so expensive per circulation that patrons will blow through their e-book budgets by “borrowing” them. Library platform providers like OverDrive, Freading, and hoopla digital may even offer libraries options to limit the number of “lends” per title during a given time period — in other words, to bring back the waiting lists they have now under PIP.

More generally, the CPC model brings libraries closer to paid subscription models like those of Scribd, Skoobe, or Amazon Kindle Unlimited. But publishers — to the extent their author contracts enable them to license into any of these services at all — will likely see them as different distribution channels and make different content available to each. They’ll likely follow the example of Hollywood studios and television networks, which have gained experience from different Internet on-demand video models (Netflix, Hulu, iTunes, Amazon Prime, pay-TV operators, etc.) and deliberately choose which movies or shows to make available over which services and at what license fees. Is this really what libraries want?

There are many obvious reasons why libraries should offer e-books to patrons, and most of them do so now. Arguments between libraries and publishers over title availability and pricing will go on for a long time, and when it comes to CPC, we’re in very early days. But when libraries consider which models to adopt — be they PIP, CPC, other models (such as OverDrive’s demand-driven acquisition), or a combination of them, if such a thing is possible — they will need to consider their traditional functions of curation and community service as well as the bottom line.