The Scholarly Kitchen will be making an appearance at this year’s Frankfurt Bookfair. On Wednesday, October 16, we will host a panel discussion (and a free breakfast!) beginning at 8:45 AM in Hall 4.2 at the Academic & Business Information Stage. At last year’s Bookfair, we discussed “The Great Acceleration”, and how the pace of change in scholarly communications was increasing. This year, we’re looking at the results of that rapid pace – what has changed and what is reaching a tipping point toward change.

As a preview of the session, my thoughts will be about a threshold I think the community has reached, largely due to the remarkable success of Plan S, the attempt to accelerate the transition of scholarly publishing to open access (OA). Although it remains unclear how well Plan S will work for researchers funded by Coalition S, it is increasingly clear that even before it has gone into effect, Plan S has achieved one of its major goals, changing the conversation around OA. What I largely hear from the community is no longer, “eventually things will move to OA,” but instead a sense of urgency, “we’re on the clock for a move to OA.” That’s inspiring a huge amount of analysis on business models, and finding ways to make that transition in a sustainable manner. Through that process, perhaps one of the most significant shifts we’ve seen is the increasingly widespread recognition that the author-pays model for OA is an evolutionary dead end.

Author-pays OA remains a growth area for many publishers (essentially a new revenue stream that comes directly from researchers and funders rather than libraries) and so far at least, the most sustainable model available, but its flaws are becoming plainly evident, and concerns about shifting the entire scholarly communications economy over to a system of Article Processing Charges (APCs) are looking increasingly urgent. In the last two weeks alone, there have been three presentations from PLOS (a pioneer of the APC model) focused on the subject of moving past the APC.

The flaws in the author-pays business model for journals have been evident for more than a decade. The APC model represents a lateral move in terms of access, greatly improving access for readers but shifting the inequity in the system onto authors. It allows everyone to read the work of others, but limits the ability to publish one’s own work to those with sufficient funds to cover the costs of doing so. This greatly disadvantages authors from less-wealthy regions of the world, along with unfunded researchers, and entire fields without the significant funding structures found in some of the sciences that are largely driving the move to APC models. Waiver programs are already under strain, and there seems little willingness among wealthy nations or authors to overpay in order to support their less-well-off colleagues.

As a publisher, the author-pays models causes a shift from focusing on readers as your customers to focusing on authors. This creates pressure to favor quantity over quality, and rewards bulk publishing approaches, as well as creating a now-thriving industry of predatory scammers. Further, APC models favor scale, and work to the benefit of the largest commercial publishers over smaller, not-for-profit, and community-owned publishers. Ask any publishing consultant or acquisition editor at a large publishing house and you’ll hear about the enormous flood of requests for proposals coming in, as previously independent research societies realize they can no longer make it on their own.

Another major issue is that library budgets are already strained and APC models are beyond the current capacity of many institutions. When the author pays, costs are concentrated solely on that author or their institution, rather than being spread more broadly across the readership of the article. This means that the institutions that largely read the literature but that don’t produce a lot of papers (“Read” institutions) will see great cost savings, while productive institutions that publish a lot of papers (“Publish” institutions) will see significant cost increases. No one knows how those cost increases will be covered, beyond hoping that funders will come up with more money, although this most likely means that publishing funds will come out of existing research budgets.

While all of this has been repeatedly discussed since the days of the first APC-driven journals, the reality of relying on author-pays models has finally begun to sink in. To their credit, many of the organizations and individuals who pioneered APC models have evolved in their thinking over time, and there is increasing consensus that we need to move beyond the APC to more equitable and sustainable models for supporting OA (e.g., Ashley Farley of the Gates Foundation: “I do not think that APCs are the solution and I hope that we move beyond them.”).

This is not an easy problem to solve – if there were obvious and easy models to support OA publishing at scale, we’d have moved to them years ago and be much further along than we are today. Like many others, I’ve been spending a lot of time thinking about different approaches, and asking theoretical questions like, if we had no scholarly publishing at all, and were going to start a system from scratch, knowing what we know today, what would that system look like? Every idea I can come up with runs up against roadblocks that I can’t yet figure a way around.

Decentralized Funding and Decision-Making

It’s often said that there’s enough money in the system already to support a flip to OA. The problem, to paraphrase William Gibson, is that the money is not evenly re-distributable.

European nations, in particular, have played a leadership role in the drive to OA. Often, these countries have a centralized funding and decision-making system in place for education and research. The universities, their libraries, and their researchers are funded by the national government, and as a result, there is essentially one set of policies and one pool of money that can be moved around to make author-pays models work. The subscription funds saved by a Read institution in an OA world can readily be shifted over to a Publish institution to cover the increased costs their researchers would face. This ability for one governing body to set publishing policy and to pool and distribute publication funding opens up intriguing possibilities for new models.

There is something deeply compelling about systems like SciELO and AmeliCA, where community resources are pooled and then distributed to a variety of not-for-profit publishers who are required to meet certain standards. But I struggle to envision how such a model could drive innovation and reward novel approaches that excel. We’re not seeing the same level of technological innovation and methodologies coming from these programs as we see inspired by the competitive market.

Perhaps a move to a pared-down, yet stagnant, publication system would be acceptable, with journals providing the basic services required and nothing further (peer review and a PDF). The entrepreneurs and startups will move elsewhere, and the technological innovation, new metrics and tools will move to different stages of the research workflow. It certainly fits with the current business strategies of the largest commercial publishers.

That may not be an insurmountable problem, at least when compared to the question of pooled funds when one starts to think about countries that don’t have centralized education and research funding systems. The United States is the prime offender here, responsible, according to the latest STM Report, for 41% of the global revenue in the existing system of scholarly communications.

To be clear for our non-US readers, there is no centralized, federal higher education system in America, no US government-funded and controlled set of universities. Instead, our public universities are funded and controlled by the 50 individual states, and these are supplemented by a large number of private institutions, funded by tuitions, donations, and their often significant endowments.

That means there is no centralized pool of money that can be shifted around to accommodate new communication models. If a Read university in Arkansas recognizes cost savings through OA, the funds provided by the taxpayers, tuitions, and student fees of Arkansas are not going to be sent to help cover increased costs for the poor researchers of Harvard (with its $39 billion endowment), a Publish institution that will see its costs skyrocket. The public education system in America is increasingly underfunded. Any sudden influx of funds (or freeing up of previously-spent funds) will rapidly find a new purpose within a given university. And that brings us to the other major roadblock to new models:

Free Riders

At the recent Society for Scholarly Publishing New Directions Seminar, one of the main proposed models for moving beyond the APC was Collective Action, that is, libraries voluntarily choosing to provide financial support to services they ideologically agree with, but could instead receive for free. While this approach has seen some success, at least on a small scale through programs like the Open Library of the Humanities or community support of arXiv, it is unlikely to be broadly sustainable in the long term. Again, this comes back to how libraries and universities are funded.

Any system that is under great financial strain and that allows for free riders is going to be largely populated with free riders.

In the US, there are state laws preventing the spending of public funds on things that are available for free or for making charitable contributions. Even in states lacking these prohibitions, university funding levels are simply too low to enable large scale unnecessary spending. In the Twitter discussion of the New Directions Seminar, I posted the following quote from Gwen Evans, the Executive Director of the OhioLINK Library and Information Network:

The problem I see with other proposed voluntary OA subventions is that under the pressure of decreasing budgets in higher education generally, but especially in public institutions, administrations will not continue to pay for something that they will receive for free. It’s a nice idealistic hope, but administrations often cut budgets for services and content that are, or were until the moment of the budget cut, perceived to be highly valuable or necessary. Any model that relies on substantial voluntary support of OA (whether OA provided by current publishers or a hypothetical shift to library/institutional owned and operated OA publishing ventures) is highly suspect from my observation of state funded education in the state of Ohio. The financial mantra is efficiency and cost effectiveness; voluntary financial support for content that is free is neither.

The responses largely echoed the reality of library funding:

This is not a "knee jerk" reaction. This is the reality that we libraries have to justify our budgets and we will be asked why we are paying for something that is free.

(FWIW, this is going to be asked of us as WTF are you paying for something free?) #sspND2019

— Lisa Hinchliffe (@lisalibrarian) October 2, 2019

Related: I have been asking a lot of librarians at *read* institutions whether they would continue to pay into a consortia deal if everything flipped to OA (or mostly flipped to OA). Unequivocal answer: no. There is a 0% chance of universities that don't *have to * pay paying.

— Michael Clarke (@mtclarke) October 2, 2019

1/2 And even if the procurement officers don't, you can be certain provosts and other fund-allocation officers will. The question they'll ask isn't "Is the free thing we're paying for a good and righteous thing that deserves to be funded?" but rather…

— Rick Anderson (@Looptopper) October 2, 2019

2/2 …"Why are we funding this particular good and righteous (but free) thing when we don't have enough money to provide scholarships to all deserving students, or to refurbish labs and classrooms, or to provide mental-health services?" Budgets are, unavoidably, a zero-sum game.

— Rick Anderson (@Looptopper) October 2, 2019

Any system that is under great financial strain and that allows for free riders is going to be largely populated with free riders.

There are no easy answers here, and no models available without significant negative consequences. This is the biggest complaint I’ve heard about Plan S – not the desire to move to OA, but rather the required immediacy, despite the lack of a clear roadmap to a sustainable and equitable future. Those complaints are not just coming in from traditional established publishers like scholarly societies, but also from born-digital, born-OA publishers, who are already seeing themselves shut out of library budgets in the “transformative agreement” gold rush (note: you can’t sign a “Read and Publish” deal if all of your content is already OA). The uncertainty is pushing the market into the hands of the largest and most-profitable publishers, those that can withstand potentially troubled times ahead.

And so we see most publishers making a good faith effort to accommodate Plan S, but doing their best to contain the potential downsides of the APC model, which at the moment is the only route available that offers some chance of staying afloat, or at least an ability to hang on until new models arise. Will that happen before the roof caves in for most? It’s hard to say, but at least we have reached a point of urgency where great efforts are being made to conceive of and experiment with new models. All of the energy that used to going into arguments about “should we?” is now going into questions of “how can we?” and this is important, as the survival of the many things the community values about scholarly communication depend on finding viable new routes.