Under read-and-publish agreements, articles aren’t deemed eligible for open access publication until they are accepted by the publisher, leaving authors frustrated, libraries stuck, and everyone needlessly confused. There’s a better way.
Imagine this scenario: You are a librarian who has spent months negotiating an uncapped open access (OA) agreement with a publisher. You go to great lengths to advertise the agreement on your campus, and authors are excited to publish in the publisher’s journals. An author submits an article, but 18 months pass before it is accepted. In those 18 months, your budget is cut by 20 percent, and you can no longer afford to renew the agreement. When the author gets their notification that their article has been accepted, it comes with an unexpected bill for $3,000.
OA agreements are popular; the University of Minnesota Libraries have participated in several since 2019. We try to stick to OA agreements that align with our open access values, so the majority of our agreements support Subscribe to Open (S2O) journals and open access monographs. Additionally, all but one of our agreements have been uncapped (meaning there is no limit on the number of articles covered by the agreement); given the problems capped agreements present for authors and libraries, we won’t agree to them going forward. We do, however, participate in read-and-publish agreements with a few large publishers through our consortia.
One troublesome aspect of these agreements is that the publisher’s determination that a manuscript qualifies for discounted or free open access publication occurs at the point of acceptance, not the point of submission. While this might make sense at first glance, it ultimately causes several problems, particularly when an agreement is up for renewal, and especially if the negotiations are protracted and/or contentious.
In our roles as director of collection strategy and e-resource management and scholarly communications librarian, we’ve been asked by many researchers if they should submit their manuscript to a publisher with which we have an expiring agreement, or whether their submitted article will still be covered if we no longer have an agreement when it is accepted. This puts the library in an awkward position, especially if we want—or need—to walk away from an open access agreement.
To alleviate author confusion and frustration and to create a fairer renewal process—one that does not trap institutions into renewal—we propose that author eligibility should be determined at the point of submission. Not all submissions result in an acceptance, but basing the OA guarantee on the submission date would provide clarity for all parties that if the article is accepted, whether six months or 18 months from submission, it will be published OA without any additional fee.
Timing
The publishing timeline is a long one—sometimes extremely long. There are many reasons for this: the process itself is time-intensive and it is impacted by internal and external factors, both within and outside the control of authors and publishers. One process over which authors have no control—nor, so it seems, do publishers—is peer review.
It is difficult to get data about how long peer review takes. One systematic review looking at 69 studies of the time from submission to publication for biomedical journals found mean and median time spans ranging from 91 to 639 days and 70 to 558 days, respectively (Andersen et al., 2021). For agreements that UMN participates in, publishers typically do not include the date the author submitted the article and date of acceptance in publication reports. But in looking at the “received” and “accepted” dates for a sample of articles covered by our agreements, we found many cases in which the acceptance date was more than six months after the received date and relatively few in which the period between submission and acceptance was less than two months.
This lag means that we cannot guarantee that an author who submits an article to a fully open access journal under a read-and-publish agreement—even six months before the end of the agreement—will be covered by the agreement. Libraries can’t in good conscience promote an agreement that might end before an article is accepted, particularly when we are not in a position to pay individual fees for authors.
Improved Communication
Switching the basis for eligibility from acceptance date to submission date would make communication with authors easier. From emails they have sent us, we know that authors often submit articles without paying attention to the terms they’ve agreed to, and that they are sometimes shocked when the article is accepted and they are asked to pay a fee. Although we’ll likely never be able to completely avoid such a scenario, tying eligibility to submission date would at least reduce the number of authors who receive an expensive surprise when their article is finally accepted.
While authors should pay closer attention to fee requirements, the messages they see when they submit articles do not always make it clear that their article even could be ineligible for open access publication if the peer review process extends beyond the length of the agreement.
For example, we previously participated in an uncapped agreement with one fully OA publisher. When submitting their manuscripts to this publisher, our authors were presented with a dropdown indicating that our institution had an open access agreement and that there would be no APC charged. In the next step in the submission process, however, the author was asked to agree to pay the APC (or, if they had no funds, to apply for a waiver); the full dollar amount was even listed (e.g., $2,400).
This was confusing; the author had just been informed that they were covered by our agreement. So why ask them to agree to pay an APC? In case their manuscript was accepted after our agreement expired. But most authors were not aware of the end date of the agreement, and nowhere did it mention that their article must be accepted before a certain date.
The publisher in this example has a long peer review process: our data from 2024 shows an average of 126 days, ranging from 36 to 265 for original research and review articles. On average, to be covered by the agreement in the year it expires, the author would have to submit their article by the end of August and potentially as early as April.
Improved Relationships between Publisher and Library
Switching the OA eligibility date to the point of submission would improve the publisher-library relationship. When authors don’t have clarity about what to expect, the manuscript submission process can result in frustration; when authors are frustrated, the library-author relationship suffers, and thereby the library-publisher relationship suffers. On the other hand, by changing the timeline of OA eligibility, publishers would demonstrate their willingness to be good partners to libraries.
For libraries considering whether to renew an expiring OA agreement, articles that are awaiting acceptance—and the specter of authors angry that their articles were accepted after an agreement ended—can be a source of immense pressure; it’s hard not to wonder whether publishers are intentionally moving slowly on articles close to an acceptance decision. This may be an unlikely scenario (and we have no evidence to suggest it is happening); but basing OA eligibility on the point of submission would eradicate all appearance of impropriety.
Increased Equity for Authors
For authors submitting to fully open access journals, the lapse of an open access agreement has material consequences. If the author loses eligibility, they may end up with a large bill months after they submitted their article, at which point the opportunity to request a waiver for the APC has passed. By this time, they may have run out of grant funding, or their grant may have ended, along with any funding that could cover the APC.
Aligning eligibility with submission also allows authors (and peer-reviewers!) to cope with unexpected life events. If an author is diagnosed with an illness, has to take care of a sick relative, or has their house burn down, they may need more time to resubmit an article after peer review; but experiencing such an event should not mean that the author becomes responsible for paying a several-thousand-dollar bill.
Publishers could argue that authors who know that an agreement is ending may try to game the system by submitting more articles to try to beat an eligibility deadline.
This is unlikely to happen. Faculty aren’t gaming the system now. Those even aware that they are participating in an OA agreement likely think their eligibility is based on the article submission date. Researchers tend to be more focused on internal deadlines related to conference, grant application, and promotion and tenure dossier submission dates, making it unlikely that they would change their practices to take advantage of an agreement where eligibility is based on submission date.
Besides, authors can only write so many articles so fast. Even if an author wanted to increase their pace of submissions as an agreement was winding down, they probably couldn’t.
All future uncapped OA agreements should use submission date instead of acceptance date, so the money paid during a given year supports OA publishing for articles submitted during that year. This clarity would make OA agreements easier for libraries to explain to authors while making them more functional and improving the publication process. This shift would also make working with a given publisher more attractive to libraries, a competitive edge in stressful budgetary times. Publishers should recognize there are few downsides to using submission date instead of acceptance date. During negotiations, libraries should insist on it.
Andersen, M. Z., Fonnes, S., & Rosenberg, J. (2021). Time from submission to publication varied widely for biomedical journals: a systematic review. Current Medical Research and Opinion, 37(6), 985-993. https://doi.org/10.1080/03007995.2021.1905622
10.1146/katina-071525-1
