Almost since its inception, the pros and cons of an Article Processing Charge-based (APC) business model for Open Access (OA) publishing have been described and discussed by the many stakeholders in the scholarly communications ecosystem. Within the last couple of years, publishers have started to experiment with other models for funding open access, and the roadblocks that exist in funding open access have been scrutinized to some extent as well. Even this week we had a guest post considering the future of OA business models!
So this month we asked the Chefs to think beyond the APC and consider: What‘s next for OA?
Rick Anderson: I don’t like rhetorical manipulation, and for that reason I’m always skeptical of calls to get “beyond” something — no matter what it is. I find that word manipulative in this context because it implies (without going to the trouble of actually arguing) that the thing in question really is something we need to “get beyond.” In this particular case, it seems to me that before we can talk about how to get “beyond the APC,” we need to establish whether doing so is necessary, or even desirable.
And I’m not convinced that the answer to that more fundamental question is necessarily yes. The APC funding model has serious downsides, of course. It inevitably discriminates against authors who have less institutional support and against fields of scholarship for which such support is scarcer. It shifts costs rather than necessarily reducing (let alone eliminating) them. It threatens to redirect money from the support of original research. And, perhaps most egregiously, the APC creates an unavoidable conflict of interest for the journal in question — not an insurmountable conflict, but one that must be recognized and dealt with. But the APC model has upsides as well. Chief among them the fact that it creates a directly scalable and sustainable model for funding open access (at least at the individual journal level). Compared to other funding models, such as the institutional subvention and the various kinds of crowdsourcing approaches, it creates fewer inefficiencies and fewer risks of unsustainable levels of free ridership.
So the APC model is a mixed blessing. This is something it has in common with every other funding model available to support OA — and with every other model of publishing, period.
So the APC model is a mixed blessing. This is something it has in common with every other funding model available to support OA — and with every other model of publishing, period. What this suggests to me is not that “what’s next for OA” will necessary mean getting “beyond the APC,” but rather that the future of OA will almost certainly be diverse rather than monolithic. Some models will work better in some publishing contexts, and in some global regions, and for some disciplines, and in some political environments, than in others. To me, this seems like an argument against trying to impose a single vision of scholarly publishing on the entire world. Why not use APCs where they make sense, and institutional subventions where they make sense, and maybe even (hear me out, now) allow for toll access to persist as part of a flexible, responsive, and highly diverse scholarly communication ecosystem? In other words, why not encourage diversity? I would suggest that what we really need to “get beyond” is the idea of totalizing solutions.
Charlie Rapple: To bastardize a phrase: OA is dead; long live OA. OA as a topic of debate is dead, because there is sufficient consensus that it is for the good, and worth pursuing. And thus “long live OA”, because it has been adopted as policy by enough major / influential funders that there is no doubt that OA will progress to being mandated by more and more governments and funders, at least in the fields (or regions?) where a sustainable business model can be made to work.
There are a hundred nuances to take issue within this opening premise of mine, but I think all of this is broadly true. One approach, then, to answering Ann’s question would be “what sustainable business models are there, if we try and take OA beyond APCs?”. But where you stand depends on where you sit, so given my interest in research communications and engagement, I’m going to interpret the question differently. Or, in truth, answer a different question – “Beyond OA: what’s next for open?” or possibly even “Beyond OA: what’s next for funders?”.
Making something openly available doesn’t make it openly accessible, in the sense of being able to be found and understood by audiences outside academia.
In both cases, my answer builds on a definition of OA as open access to scholarly publications, which again, is probably the most common use of that specific term, as opposed to “open science”, “open data”, or “open” more generally. Hence I make the distinction between OA and open, which – as I’ve argued so many times – is an important one. Making something openly available doesn’t make it openly accessible, in the sense of being able to be found and understood by audiences outside academia. We already see the implications of this distinction in the work of organizations like Hindawi and PLOS, investing in lay summaries and other statements of broader impact, that help to maximize the open value of the work beyond the ’simple’ business model of open access.
And this relates, too, to my second interpretation of the question, as funders are also increasingly requiring researchers to articulate in simple terms the potential impact of their work, and to open up their approach to impact and ensure “real world change” as well as advancement of knowledge. In a research project earlier this year (which I wrote about in the Kitchen here), it was telling to see that broader impacts and knowledge exchange (strategic efforts to engage non-academic partners with research projects and results) now rank more highly than open access publishing, sharing of data and public access deposit, in terms of funder requirements for grantees.
So that, from where I’m sitting, is “what’s next for open” — growing expectations from funders around “open communication” of research to ensure that its potential impacts are broadened and maximized.
Alison Mudditt: I often find myself returning to the roots of PLOS and the core goals of our founders: fast, efficient and complete access to the scientific record for all in order to accelerate the pace of discovery. To me, that last phrase is key – our work towards open is in service of a higher and more consequential goal. Viewed in that light, we’ve had some very significant wins but there is much work still to be done to establish a landscape for open research that is sustainable, equitable, and globally inclusive. For me, there are two key priorities that stand out – and both are a focus for us at PLOS.
We must expand business models beyond the APC. I’ve been saying this for a long time – since before I arrived at PLOS: while APCs have demonstrated the power of an OA publishing model and that OA can create value as a business, they erect a participation barrier for far too many. We now have the opportunity to help push OA forward to a knowledge future that is fairer and more accessible to all. There are plenty of roadblocks (summarized in an excellent recent post here in the Scholarly Kitchen) but the importance and urgency of this issue needs to be acknowledged by those of us eager to see a fully open research landscape. It’s encouraging to see growing innovation here but I worry that other strong pressures are pushing us headlong towards a gold APC world that won’t serve our core objectives well.
It’s encouraging to see growing innovation here but I worry that other strong pressures are pushing us headlong towards a gold APC world that won’t serve our core objectives well.
We need to move beyond OA to a landscape that facilitates, incentivizes and rewards the full range of open research practices. There is no single silver bullet solution, and we’ll need both major breakthroughs and a multitude of smaller advances that together move us toward our ultimate vision. This is about a much wider behavioral and cultural transformation from a system that focuses far too much on where researchers publish to a focus on the research itself and how it both advances their fields and contributes to society more widely. A system of scientific discovery and communication that recognizes and rewards the few at the expense of the many will not serve us well as we face a multifaceted global crisis that includes soaring inequality, the rise of populist demagogues and the climate crisis.
Sian Harris: Although APCs have helped the system adapt to enabling greater access, in many cases, they have simply shifted where the inequity lies rather than removing it. In a survey of researchers in INASP’s AuthorAID network by a colleague and me (preprint here), we found that many researchers in the Global South still pay APCs, sometimes themselves, suggesting gaps in awareness of or eligibility for publisher waivers. We have also found that sudden changes in waiver conditions when a country’s economic status improves can be challenging; a multi-year research grant cannot easily accommodate a jump from zero charge to, say, 50% of an APC that could run into a couple of thousand dollars. In addition, requiring researchers or their institutions to ask for APC waivers can perpetuate global power imbalances.
It is therefore encouraging that conversations about funding OA are moving away from current approaches to APCs in response to Plan S and it is encouraging that the issues I mentioned in the previous paragraph are being recognized in a range of conversations around Plan S. To really enable equity, which is the theme for this year’s OA Week, these conversations and decisions need to not just recognize the implications for researchers and institutions in the Global South but also to include Southern researchers, librarians and others in the discussions and decisions.
Another important aspect to bear in mind is that journal publishing is much wider than large commercial Northern publishers and that not all journals take the same approach to APCs.
Another important aspect to bear in mind is that journal publishing is much wider than large commercial Northern publishers and that not all journals take the same approach to APCs. Many journals in the Global South charge very small or no APCs, as fellow Chef Haseeb Irfanullah discussed in this article in relation to Bangladesh. There are potential lessons to be learned more widely from models where funding comes from research institutions or government bodies and is not tied to specific papers or researchers.
David Smith: A couple of years back, I was invited to a weekend workshop in order to to kick about some ideas about what publishers could do to go beyond the current paradigm of scholarly research outputs. The non-publisher participants — librarians and funding agencies — were somewhat surprised at the level of engagement by us publishers. But then came the problem… No extra money. A big wish list of exciting ideas requiring infrastructure and processes and workflows. But the same pot of cash. So I asked what would need to be given up in order to free up funds for these ideas.
So what’s next for OA? Well. There is a world where the wishes and desires are costed and modeled and the funds are made available (however that happens), and then the infrastructure needed to support that new world is built, and is stable, and the business of scholarly information sharing, curation, validation, and promulgation changes to a new paradigm (again however that happens). But that won’t happen off the APC while it’s stuck in a world of identity politics and all the other nonsense that seems to accompany it.
Personally, I’m a fan of Object Oriented Publishing where the research is broken down in to modules and reassembled based on the the requirements of the person or machine agent seeking a particular informational chunk.
Personally, I’m a fan of Object Oriented Publishing where the research is broken down in to modules and reassembled based on the the requirements of the person or machine agent seeking a particular informational chunk. But that world isn’t cheap to build or maintain, let alone grow. So I suspect the real answer is “not much”. But then I’m typing this as my country marches and our Parliament ‘debates’ the sorry milestone of an ideologically driven idea that at no point has been properly thought out and understood in terms of its long term consequences. OA has always been a wonderful idea. But it badly needs to work on the detail of what ‘free at the point of use’ actually entails in the networked age.
Alice Meadows: My somewhat tongue-in-cheek answer to this question is that it isn’t the right question to be asking. Because whenever someone asks what’s next for open access — never mind some specific aspect of it — I can’t help thinking that the real question is what’s next for open research. For me, what we can potentially achieve through a more open research process, from start to finish, is so much more important (and interesting!).
Of course research results should be publicly accessible in some form, but that applies to the data and other supplementary materials just as much as the research paper. What about open source code — isn’t that just as valuable?
Then there’s the fact that making content open access doesn’t necessarily make it more easily discoverable or, therefore, used. A friend shared a great example of this recently. She works for a not-for-profit organization that routinely makes all its publications openly available, and yet usage was historically very low — until they started using open identifiers (specifically DOIs), leading to a significant increase in the number of downloads. Open access alone wasn’t enough.
And let’s not even get started on the benefits of opening up other parts of the research process!
Instead, to (briefly) answer the actual question at hand, I don’t think APCs are going to disappear any more than preprints, author-accepted manuscripts in repositories, or any of the other existing forms of OA. It’s a mixed bag and, I’m sure, will continue to be.
Instead, to (briefly) answer the actual question at hand, I don’t think APCs are going to disappear any more than preprints, author-accepted manuscripts in repositories, or any of the other existing forms of OA. It’s a mixed bag and, I’m sure, will continue to be. As it should, given that in scholarly publishing, as with everything, there are no one-size-fits-all solutions. Different disciplines, publishers, and other communities will always need different approaches to OA. So while read and publish/transformative agreements may have taken over from APCs as flavor of the month right now, I find it hard to believe that they will ever be the only approach. Funders and others will surely continue to seek ways to reduce the cost of the publishing process and publishers will continue to find innovative new ways to get around any restrictions that are placed on them!
As one would expect, the Chefs have varying views on what might be next for OA. Several have pointed out that OA is only a small part of open research or open science and that what we should be considering is what is next for “open” more broadly.
Now it’s your turn.
What do you think is next for OA? What might lie beyond the APC?
22 Thoughts on "Ask The Chefs: Beyond the APC"
Reading the responses brought to mind this post from 2013:
We’ve been arguing for 20 years about access to the stories written about research results, rather than access to the results themselves.*
*at least in the sciences.
The funder s will fund until they either can fund no more because there are no more fundable funds or get tired of funding and move on to something else. In either case someone will say: Hey, why not let the users pay for what they want to read. How about a subscription based model? As the old saw goes: What goes around comes around!
Before we talk about how we get to “free” we should first discuss what we will get for “free”. Because the more you remove the profit motive the less desire there is to do the work. In pursuit of cutting costs to the authors and the readers we have already seen serious journals do away with copy-editing and proofreading, and sometimes even uniform presentation, and, I believe, we have seen a decline in the quality of the papers being published, both in terms of overall presentation and in terms of the quality of the science being reported. Because the cheaper we make it to submit, and to publish, and to read, the more we encourage submission of subpar, poorly written, poorly presented work which is then accepted and published under the idea that more is always better.
In a perfect world, Bill Gates would take his billions and fund long-term endowments which would allow continual publication of journals which apply rigorous peer-review and strict publication standards. But that wouldn’t be “free” and Bill has chosen to do other things with his money.
See also “Ramifications of the Downward Pressure on Pricing”
I scanned through the chefs’ offerings perhaps too quickly, but could not see any comment on the role of research funders in meeting the costs of publication. How can we move towards a world where the costs of publication are routinely covered as part of the costs of the research? Communication of research should (where possible) be part of the research budget.
This, of course, would solve one problem (the burden of APC payments on authors) by creating another one (the redirection of research funding away from research and towards dissemination). The more grant funding gets dedicated to covering publishing costs, the less actual research it can support.
Please note that I’m not asserting that this trade-off would be bad; I’m only pointing out that it’s a trade-off.
This would also remove a check on the work of funding agencies, as they would in effect be both funder and arbiter. Publishers provide independent critique of the work of funding bodies.
The APC model places the value on publication, and the associated costs are covered entirely by the small number of authors, their organizations, or indirect subsidies from outside agencies. The significant value to readers is ignored, and the traditionally requiring subscription subsidies are removed. This long tail of reader support funding (or some sort of reader/society support) is necessary in order to create a viable business model – unless the processes and costs of publication are dramatically reduced. Moving back toward reader/community support, based upon reasonable costs and altruistic goals, creates a less gameable model, and reduces the tendency to have private and society publishers strive for maximum profit – rather than maximum exposure. There are a number of ways to funnel community support into efficient publishing models…all of which should separate and minimize the costs of both peer review and distribution. These models have been discussed for a long time; for a few of my ideas reaching as far back to 1999 see: “Pricing Models: Past, Present, and Future?” Serials Librarian 36 (1/2): 301-319 (1999) https://docs.google.com/file/d/0B9eTeOGYv-D2aG9yT2k2R25MYzQ/edit or my recent column on distraction concerns at “Open Access: Misconceptions and Misdirections.” Against the Grain 31 (2): 36. (April, 2019) https://drive.google.com/file/d/1jT4bt-JcyLHP2uCYfMf2nGUmgy1oisTy/view?usp=sharing
“Moving back toward reader/community support, based upon reasonable costs and altruistic goals, creates a less gameable model, and reduces the tendency to have private and society publishers strive for maximum profit – rather than maximum exposure.”
Societies were created for the most part (many in the 1800s) to have a platform to publish a journal. The origin of journal publishing was to provide a forum for scholars and researchers to share their work with each other, not for for-profit companies to make a profit off researchers’ sweat (and authors and editors and reviewers free work).
“Members of those societies pay membership fees that support their journals . . . ”
I have worked for a number of journals that never had APCs in their history and that transformed to OA by dedicating membership funds and by asking for donations for OA specifically (who doesn’t want open scholarship?). When nonprofit societies started getting all corporate and decided each “program” of the society had to pay for itself (impossible), they were getting away from the original purposes and mission and goals of their society. Then they make exceptions, such as student scholarships to the meetings can come out of membership, donations, meeting registrations, etc., so there goes each program paying for itself. Unfortunately, some small (and larger) societies look at the Big Deals and say shouldn’t we be making some big bucks, too. Just wrong thinking for their organization.
“Societies were created for the most part (many in the 1800s) to have a platform to publish a journal. The origin of journal publishing was to provide a forum for scholars and researchers to share their work with each other, not for for-profit companies to make a profit off researchers’ sweat (and authors and editors and reviewers free work).”
Societies were created to allow people with common interests to connect. The journal was a means to that end. Not the end in itself. Many societies do not have a journal as they use other means to enable their members to connect.
Oldenburg started the Philosophical Transactions (the first scientific journal in English – and still going) as a private venture with a view to making a profit. The Royal Society (the world’s first Society, formed before the journal concept existed) only took it on after his death (he found making money from it tricky).
I suggest it’s poor form to rewrite history to fit modern prejudice.
I was thinking of American science societies and should have said so. I hope there is still a place for nonprofit societies to publish journals, make them open access, and be a contributor to science publishing from their own point of view.
I’m confused by all of this. It seems simple to me: either the reader pays, and/or the author pays, and/or a philanthropist pays. Isn’t the question: which of those, or what combination of those enables an economically viable publishing ecosystem? Joe Esposito has pointed out that publishing a paper costs ~ $5K (on average, industry wide; ). It seems to me the only question is who pays that. If it’s the authors or a philanthropist then it can be OA. If the reader makes a contribution it can’t be OA. Am I missing something?
Mark, that’s not quite what I said. I said that publishers receive, on average, $5,000 per article published in the traditional model, figure I picked up from Andrew Odlyzko. That’s based on a $10 billion industry and 2 million articles/year. If the industry makes money overall (no one knows: it’s not all Elsevier and Wiley), then the cost *on average* per article is under $5,000. If the industry loses money, then the average cost per article is greater than $5,000. It could be (this is simply a speculation) that the industry averages a profit of 15% *on average.* If so, then the average cost to the publisher per article is 85% of $5,000. Of course, it costs NEJM far more per article than a small, relatively unknown publication. Averages can mislead.
The “average” thing is increasingly problematic. Policy makers seem to look at an “average APC” or some other average number, and declare that as the amount they intend to pay (or the top limit of what they’re willing to pay). This ignores the nature of what “average” means, that half cost less, and half cost more, and seems a poor way to come up with an equitable payment system.
The $5,000 per article “cost” is not an accepted industry standard — it is what one/some commercial publishers claim is reasonable. This costs assumes there must continue to be a healthy profit margin. Non-profit editors using OJS (free) software have estimated a value of closer to $900 per article. Of course there are scales of economies issues with large and prestigious journals, but separating and making peer review costs and distribution costs transparent would also help the industry come up with a better minimum cost per article continuum. Such evidence-based costs would be a better basis for a conversation about the real and viable costs of APC and alternative models.
Sometimes the “philanthropist” who pays for open access is the Society that publishes the journal, and they do that through their membership fees (and donations). This is perhaps the simplest way for a journal to be 100% OA. Read and Publish deals sound awfully complicated, and perhaps are created to prolong the inevitable move from partial OA to full OA for a journal (giving time to the for-profit publishers to figure out what business to get into next since they don’t seem to be able to embrace OA fully).
If the “philanthropist” paying for a Platinum OA journal is the members of the society through their membership fees, then that means the costs are being paid by the researchers themselves, and since very few (any?) grants will pay for society memberships, this money is coming out of their own pockets. I’m not sure there are a lot of fields with a wealthy enough population of researchers (graduate students don’t have a lot of money in my experience) to cover the costs of a high quality publication program.
I’m aware that the $5K/paper estimate includes some profit, which I (perhaps carelessly) included in “cost”, because it seems to me that profit is necessary to keep publishers motivated and in business. And I realize that this estimate has a relatively shallow distribution with long tails. But whatever the price, it seems to me it all boils down to who pays, authors, readers, or philanthropists? The more readers and philanthropists pay the more we’ll be able to get beyond the APC, but the more readers pay the less OA we’ll be able to have.
All true. But you left out one thing: the more readers pay, the more content you have.
I agree that profit is necessary to keep (commercial) publishers motivated. Perhaps that is part of the larger problem; we incentive an unnecessary element of a scholarly communication network. There are examples of successful non-profit publishers that could scale up to cover the entire network – if only we could migrate the editorial boards and reallocate the large profits to cover infrastructure costs. We could probably reduce costs in this way. One other problem with financial incentives driving the network infrastructure is that we have seen commercial publishers simply drop the archival responsibilities for certain titles when they are no longer profitable – and the previously mentioned non-profit organizations adopted them to address the actual scholarly network priorities. Perhaps we should leave profit publishing for trade materials and design the scholarly publishing network as a non-profit operation — since much of effort is already provided gratis by altruistic academicians. It would be a shame to unnecessarily bifurcate the scholarly publishing world into expensive prestige journals in grant-funded areas and less expensive non-profit journals in other disciplines. University administrators, faculty authors, professional societies, and concerned citizens could advocate for the least expensive effective model rather than for a more costly network.