The loudest document in open-access publishing this decade is not a mandate. It is the invoice that a university library sends to a publisher after a read-and-publish negotiation. Mandates generate press releases. Budget lines generate negotiating power.
Open access (OA) publishing simply means peer-reviewed research is freely readable online instead of locked behind a journal subscription. The principle is easy. The payment mechanics are not. When a research funder requires OA, someone has to cover the cost that used to sit inside a subscription. That someone is increasingly the academic library.
The mandates are real, and they are stacking
Plan S, launched in September 2018 by the coalition of research funders called cOAlition S, says research supported by its members must be openly available immediately upon publication from January 2021, without an embargo and under a Creative Commons Attribution (CC BY) license. The rule applies to UK Research and Innovation and the European Commission. It did not quietly fade.
In the United States, the White House Office of Science and Technology Policy released what is commonly called the Nelson memo on August 25, 2022, after then-director Alondra Nelson. It directs federal agencies to make peer-reviewed publications from federally funded research publicly accessible without an embargo by December 31, 2025. The old National Institutes of Health policy, in place since 2008, allowed a 12-month delay. That delay is going away.
These policies matter because they attach conditions to money. A grant is the strongest lever a funder owns. But a mandate tells researchers what they must make visible. It does not tell a university how to pay the new article-level bill.
The negotiators who actually set the terms
University libraries sat across the table from publishers long before Plan S existed. The difference now is what they are buying. In a traditional subscription, a library pays to read. A transformative agreement usually bundles that reading access with open-access publishing rights for the institution's authors, so the annual fee covers both. The shorthand is read-and-publish. An article processing charge (APC) is the fee a publisher charges to make an individual article open access, and in these deals the institution often pays it centrally.
The University of California system walked away from its Elsevier contract in early 2019 after the publisher would not agree to the university's open access and cost requirements. In March 2021, UC signed a four-year agreement that covers both reading access and open-access publishing for UC authors. The details are public because UC made a point of publishing them, which remains an exception in this market. The University of California website tracks the agreement in plain language for authors.
MIT terminated its largest Elsevier journal package in June 2020. Germany's Projekt DEAL, a national consortium, signed a read-and-publish deal with Wiley in January 2019 and another with Springer Nature in January 2020. Sweden's Bibsam consortium canceled its Elsevier contract in 2018 before reaching a read-and-publish arrangement the following year.
The University of California's open access publishing guide walks authors through the March 2021 Elsevier terms without paywall jargon.
Here's the catch
Read-and-publish agreements sound like a fair trade: one payment, two things. The catch is that publishers are often collecting that one payment at near the same level as the old subscription, sometimes with annual price increases, while now recording open-access publishing as a separate growing product line. The library's budget is not being liberated. It is being renamed.
The term transformative does not always mean cheaper. It means the contract changes. Whether the total spend changes is harder to know because most deal terms remain confidential. The ESAC Transformative Agreement Registry lists hundreds of agreements worldwide, but price transparency is uneven. Many universities know their own invoice and little else.
Plan S anticipated part of this. The funder coalition generally does not support paying article processing charges in subscription journals that still charge readers and authors at the same time, a practice known as double dipping. That is why Plan S pushes authors toward fully open access journals or repository routes, a position outlined in its Plan S principles and implementation guidance. The problem is that fully OA journals are not equally distributed across disciplines. In fields where they are scarce, faculty feel the mandate pushing them toward an invoice.
Then there is the author without a grant. An APC of even a few thousand dollars is impossible for many early-career researchers and scholars in low-income countries. Some publishers promise waivers. The waiver is not a guarantee.
Hype versus reality
Hype: subscriptions will collapse and publishers will be forced into a fully open access world. Reality: the major publishing companies have spent the years since 2018 converting subscription revenue into article processing charge revenue and signing multi-year read-and-publish deals. They have not collapsed.
Hype: transformative agreements will make publishing costs transparent and reduce them. Reality: the agreements made terms more central to negotiations, but actual costs often stay hidden behind non-disclosure clauses. Libraries compare what they can obtain informally.
Hype: researchers will simply choose cheaper or free routes. Reality: promotion committees still reward journal reputation, and journal reputation still clusters in venues with high APCs or premium packages. The mandate changes access; it does not immediately change prestige.
Grade the transition honestly. Open access mandates have increased immediate public access to research. That is a measurable public good. They have not yet reduced the financial burden on universities. The existing publishing oligopoly is too effective at relabeling the line item.
The library budget's pressure point
Library acquisition budgets sit inside a wider university financial problem. When tuition revenue dips or public funding stalls, academic libraries are not insulated. They are often one of the first places a provost looks for recurring savings, because journal contracts can be large and renewable.
An internal budget fact often gets lost: the money for an APC and the money for a journal package may come from different pots. The subscription line may belong to the library. The APC may be paid from a grant or a department account. That fragmentation makes it hard to understand the institution's total publishing spend. Some universities have started centralizing APC funds precisely because they want to see the whole number.
This is the same pressure that shows up in broader campus budget reporting. Public universities racing to close tuition and funding gaps often treat recurring journal contracts as one of few negotiable line items, a pattern explored in AcademicJobs' coverage of public university tuition rises. Library negotiators are now arguing that their deals should not be treated as a passive service contract but as a financial strategy. It is a harder sell than it sounds.
Photo by Zoshua Colah on Unsplash
Rights retention is the quiet counteroffer
Some universities have stopped negotiating only about money and started negotiating about rights. A rights retention policy says the institution or author keeps a non-exclusive right to make the accepted manuscript available in an open repository, regardless of what the publisher's standard contract says. The publisher can still publish the final version. It cannot lock the research away.
cOAlition S has promoted this approach as a way to deliver immediate open access without paying an APC. Under UK Research and Innovation's policy, authors can deposit an accepted manuscript in a repository under a CC BY license, though funders do not force every author into this route. Harvard's Faculty of Arts and Sciences adopted one of the earliest open access deposit policies in 2008, and the idea has spread through institutional repositories.
For library budget negotiations, rights retention changes the balance. If authors can make their work readable through a repository, the publisher's exclusive reading package loses some of its urgency. The publisher can still charge for value-added services, but the library has a fallback that did not exist when subscriptions were the only door.
Plan S's rights retention strategy explains the legal basis for this approach.
The last word belongs to the author
Open access policy can be read as a researcher problem and a publisher problem, but for the library it is a cash-flow problem. The researcher experiences it first at submission time. Martin Eve, a scholar and founder of the non-profit Open Library of Humanities, has argued for years that public access is not enough if publishing remains organized around invoices. He has made the point that removing the paywall for readers can mean building a toll gate for authors, particularly in high-volume science. That observation remains the bluntest measure of whether a mandate is working.
The test is simple. Can a well-regarded researcher with no grant and no institutional APC fund publish in the journal their field values without paying a personal bill? In too many disciplines, the answer is still no. That is not an argument against open access. It is an argument against open access by invoice alone.