Libraries don’t “buy” most e-books the way they buy print books; they license them. That reality is at the crossroads of budgetary constraints and patron satisfaction in modern libraries. With print, a library typically pays once, owns the copy, and lends it until it wears out. With e-books, publishers and intermediaries, like plat-forms such as OverDrive, commonly offer time-limited or loan-limit-ed terms (for example, a license that expires after a set number of checkouts or after a set period), can impose embargoes, and can set prices far above consumer retail. Libraries argue these terms frustrate their public mission and make it hard to meet demand; publishers respond that digital copies don’t degrade, can be dis-tributed at near-zero marginal cost, and require a different busi-ness model to sustain authorship and publishing investment.
COPYRIGHT LAW: THE FIRST SALE DOCTRINE
The lending model of libraries in the print world was enshrined in United States copyright law in the “first sale doctrine.” The first sale doctrine1 limits a copyright owner’s control over a copy (e.g., a book or record album) after its first authorized sale. In plain terms: Once you own a lawful copy of a book, you can gener-ally resell it, give it away, or lend it without needing the copyright owner’s permission. That doctrine is a cornerstone of traditional library lending. But with e-books, libraries do not acquire owner-ship of a copy; they acquire a license governed by contract and implemented through controlled access systems. Even when a li-brary pays “as if” it is buying, often at multiples of an individual’s purchase price, publisher offerings are not structured as sales but instead as licenses. E-book licenses typically involve reproduction, transmission, and other controls in tandem with preempting federal copyright law.
HOW LICENSING SHAPES E-BOOK ACCESS
E-book licensing operates more like a negotiated service relationship than a sale. License terms for an individual e-book or e-book package typically address: (1) who may access it (e.g., authenticated cardholders, sometimes only in a geographic area); (2) how many simultaneous users there may be (e.g., one user at a time per copy, or multiple users in a costlier simultaneous-use model); (3) duration and durability (e.g., perpetual access, a length of time, or “metered access” limiting use to a set number of check-outs); and (4) technical controls (digital rights management (DRM), platform restrictions, and limits on preservation/archiving). These terms are often standardized and nonnegotiable for individual libraries. This shifts bargaining power toward large publishers and dominant distribution platforms; this is especially true for bestsellers, where patrons’ demand is most intense. Regardless of the license terms, publishers and vendors sell e-books to libraries at a cost of multiples the cost for an individual consumer; this is vastly different from in the print market, where the cost for a library is generally the same as for an individual consumer.
Publishers’ central economic concern is that library e-book lending can substitute for consumer purchases more directly than print lending, because e-books don’t go out of stock, don’t wear down, and can be delivered instantly. Libraries counter that digital lending serves equity goals (e.g., rural access, disability access, homebound patrons) and that restrictive pricing and expiration terms can force repeated repurchases that resemble a recurring tax on public access. The result is a policy fight over whether the market is functioning “normally” or whether law reform is needed to prevent one side from using contract terms to defeat copyright law and reshape what lending means in the digital world.
LAW REFORM FAILURES OF MARYLAND AND NEW YORK
Maryland enacted one of the first major state efforts to force better e-book terms for public libraries.2 The law required that if a publisher offered to license certain electronic literary products to consumers, it also had to offer licenses to Maryland public libraries on “reasonable terms” while also targeting publisher practices such as embargoes.
Publishers sued, and a federal court blocked enforcement. The U.S. District Court for the District of Maryland issued a decision declaring the Maryland law unconstitutional and unenforceable on preemption grounds.3
The core reasoning was that copyright is a federal system designed to be nationally uniform, and a state cannot effectively create its own compulsory or quasi-compulsory licensing regime for copyrighted works. In the court’s view, Maryland’s approach interfered with rightsholders’ federally protected exclusive rights, so it stood “as an obstacle”4 to Congress’s objectives. This case illustrates the central legal obstacle for state reforms: If a statute looks like it compels licenses or dictates terms in a way that conflicts with federal copyright’s allocation of rights, it risks being struck down.
In 2021, the New York legislature passed a bill5 which would have required publishers to offer libraries licenses to e-books on “reasonable terms,” with enforcement mechanisms under state law. Despite widespread support for the bill, New York’s Governor, Kathy Hochul, vetoed the bill, citing preemption by federal copyright law.6
Even though the New York bill never took effect, it became a key part of the national debate because it showed that “reasonable terms” legislation on one side could pass with significant bipartisan support, while, on the other, it could be framed as a state intrusion into federal copyright law. The New York veto and Maryland litigation together sent a clear signal: State mandates that resemble compelled licensing are legally vulnerable, and even when politically popular, they may be hard to implement without a federal solution.
After Maryland and New York, states have not stopped experimenting. One notable development is Connecticut’s 2025 law,7 which prohibits unreasonable contract terms in e-book licenses for libraries; these terms include prohibition on interlibrary loans, limits on the number of times an item can be loaned, and provisions that violate state privacy law.8 The Connecticut law includes a multi-state “trigger,” wherein the law takes effect only if other states with a combined population threshold enact similar laws.9 The trigger mechanism reflects a strategic attempt to build collective leverage while reducing the risk that Connecticut is singled out by publishers.
Massachusetts and other states have continued to consider similar bills for libraries, showing the issue remains politically live even after the Maryland defeat. These newer proposals try to avoid the specific preemption problems identified in AAP v. Frosh10 by changing the legal focus, i.e., focusing on contract clauses or state procurement instead of compelled licensing requirements.
CURRENT STATUS
Right now, most libraries are stuck navigating a marketplace where the most popular/important titles are often the most constrained and expensive, license expirations can force libraries to repurchase the same titles repeatedly, and vendor/platform dependence can limit preservation and long-term access. Meanwhile, publishers are wary of rules that could negatively affect their bottom line. Along with states focusing on avenues within state contract and procurement law which may survive preemption scrutiny, many reform conversations now have shifted toward federal options to make preemption concerns disappear.