2019 was a watershed year for progress in the transition of research publishing to open access (OA). The shakeup caused by Plan S had some time to sink in, cancellations of big subscription deals ramped up, and as I noted last October, the conversation had shifted from “eventually things will move to OA,” to instead a sense of urgency, “we’re on the clock for a move to OA.” The value of open science (increased transparency, open data, open access to research results) has become increasingly obvious during the current global health crisis. Both the positives (rapid reporting and sharing of information) and the negatives (the glut of bad science being issued as preprints and promoted via mainstream media without proper curation) are now evident, with the good generally outweighing the bad. Despite the daily evidence of the importance of shifting to an open science environment for research, the economic fallout from the pandemic is going to make necessary progress difficult and slow.

Roadblock to Belgian territory as part of the COVID19 pandemic border closures
Roadblock to Belgian territory as part of the COVID19 pandemic border closures by the Kingdom of Belgium, as seen from the German territory, in Lichtenbusch, near Aachen. Image by Turelio under CC BY-SA license.

The academic community is likely to be financially devastated by the pandemic. Roger Schonfeld’s grim overview of the funding relied upon by educational institutions makes it clear that universities are in for hard times. Tuition revenues are expected to plummet at universities (such as those in the US and UK) that rely on enrollment by international students, who often carry a significant portion of a university’s tuition load. Many domestic students are also expected to take a “gap year,” further depressing enrollments. It’s unclear what will happen at universities outside of these regions, but given recent new waves of infection globally, the likelihood of business-as-usual is pretty low.

Major universities have announced that they will continue with distance learning in the fall, which means no on-campus revenue from housing, meals, activities, and various fees, not to mention the likelihood that the students who do enroll will refuse to pay full price, given that they are not receiving the full college experience. Sharp budget cuts at universities are anticipated, and given that libraries are already undervalued and underfunded, they will be a prime candidate for such cuts.

Jisc and UK universities are trying to weather the storm by asking publishers for significant discounts on all agreements, and I’ve heard from colleagues that other major library consortia are also seeking to change the rates agreed to in existing multi-year deals. I suspect that publishers will make efforts to meet these requests as best they can, but it’s unlikely that the concessions offered will be enough to offset the expected reductions.

Research societies are also going to have a tough time. Many rely on their annual meetings to provide a large portion of their operating income. As Michael Clarke wrote, we shouldn’t expect any large meetings to happen until late 2021 at the earliest. For most, that means a two-year gap in meeting revenue. Society memberships, largely purchased by individuals, may decline as discretionary spending becomes difficult to justify during hard times. This puts additional pressure on society journals to bring in revenue to keep organizations afloat.

This creates a seemingly unresolvable dilemma – the COVID pandemic has given us a clear view of the value of open science, yet has also created conditions where implementing open science may not be feasible. We know that the transition to OA was likely to be financially disruptive, even under good market conditions, but now in a global recession (if not depression), it’s increasingly difficult to see rapid progress happening.

It has become widely acknowledged that the main model for OA publishing, author-pays APC (Article Processing Charge) Gold OA is, at best, limited in its value. It works very well in some fields, for some authors, in some geographies. We know that APCs create great inequities for unfunded authors, for poorly-funded fields of research, and for researchers in countries outside of wealthier nations. On top of this, it has also become abundantly clear that for many existing journals, a shift to an APC model at current market rates would result in reduced levels of revenue. This is going to be hard to accept for commercial publishers in a time of cost-cutting and layoffs, and impossible for research societies that have lost their other earning channels. It’s hard to justify investing heavily in new technologies and workflows when you’re struggling to keep the lights on.

One result that I expect to see accelerating is the further consolidation of the market. Last year saw an unprecedented number of independent research societies reaching out to larger publishers to discuss partnerships. A shift to APC-based OA, which benefits enormously from economies of scale, was a significant driver behind these partnerships. The American Society for Biochemistry and Molecular Biology (ASBMB) for example, noted in the announcement of their plans to move to OA that:

Ultimately, we determined that the ASBMB cannot achieve its goal of full open access alone, so we made the strategic decision to enter into an agreement with the commercial publisher Elsevier, which has experience transitioning subscription journals to full open access and the requisite technical expertise and infrastructure.

The IET similarly moved from independent status to publishing with Wiley in order to support an open access transition.

Another potential problem arises from the APC model depending on productive universities (“Publish” institutions) taking on a higher financial burden, while those universities that use the literature but don’t produce a lot of it (“Read” institutions) seeing a cost savings. The American Chemical Society notes that, “Of the 6000 institutions subscribing to ACS’s journal packages, between 300 and 500 produce 85% of published papers.” That same article quotes a SPARC report that suggests, “It remains to be seen whether ‘publish’ institutions will be able and willing to accept the radical reallocation of costs logically implied by transformative agreements.”

Add in the expected library budget cuts and it becomes increasingly doubtful that many Publish institutions will be able to increase their serial spend to cover author fees. Martin Paul Eve argues that the increases required for transformational agreements are “simply not going to be possible for many institutions, particularly at this time.” Expect the rapid pace of growth of transformational agreements, at least those forward-looking enough to actually compensate for the corresponding loss in “Read” revenue, to decrease.

Relying on authors to pay APCs out of their grant funds may seem like a good alternative to further taxing the library, but this may also be problematic as the global economic downturn is already impacting research budgets, as has been announced for Horizon Europe. This also bodes poorly for investment in communications infrastructure, data repositories, or funds to cover compliance and enforcement of open science policies.

Business models beyond the APC may have an even bigger struggle ahead. Because of the many shortcomings of the APC model, a variety of OA business models that can be applied in different contexts and that are appropriate for each community and research field are needed for long-term sustainability. Right now, most of the non-APC models in-play rely upon voluntary spend from someone. Will the cost paid for publication of a Diamond-OA journal out of a library make the cut when budgets are being slashed? Collective action strategies that rely upon libraries voluntarily paying for memberships or subscribe-to-open models are going to be similarly hard to justify, given that you receive all the same benefits of the model whether or not you choose to pay.

And so we are left stuck between a growing consensus by the research community that open science is the superior way to drive progress and an inability to invest what may be needed to make it happen.

All is not without hope, however. Though we are already seeing significant decreases in the GDP of most countries, recognition of the need to fund scientific research has never been stronger. Along with that realization, the value of open science practices is increasingly understood by governments and funding agencies. This creates the potential for continuing progress through the creation of incentives and eventually, after normality resumes, the dedication of increased funding to both support existing OA models and to drive the creation of new models. Clear requirements on researchers and grant programs for building open science infrastructure and experimenting with new models will be imperative in maintaining the forward momentum.

There are also business opportunities here, and in this case fortune will favor the bold, those willing (or able) to invest during hard times. Unfortunately, the number of players in the market with deep enough pockets to make these sorts of moves is limited, and likely to lead to further consolidation.

Open access relies on the concept that knowledge is a public good, but acknowledges that there are costs and efforts necessary to produce and maintain that public good. The global health crisis has the potential to bring stakeholders together in support of improving the way we communicate research results, but the accompanying economic downturn may create significant roadblocks to those efforts.

David Crotty

David Crotty

David Crotty is the Editorial Director, Journals Policy for Oxford University Press. He serves on the Board of Directors for the STM Association, the Society for Scholarly Publishing and CHOR, Inc. David received his PhD in Genetics from Columbia University and did developmental neuroscience research at Caltech before moving from the bench to publishing.

Discussion

16 Thoughts on "Two Steps Forward, One Step Back — The Pandemic’s Impact on Open Access Progress"

It is clear that academic library spend will decline significantly with big deals, especially with the advent on the Unsub tool, and subscriptions taking a big hit, possibly in the 25% range or more. One would expect this to hurt publisher margins and the surpluses societies take from their publishing programs. How the financial pressure on APC support effects researcher behavior will be interesting to watch. I would expect more use of preprints and consolidation of research reporting into fewer articles covering more work. However it plays out how libraries acquire content and how research communicate their work will be different in the post-pandemic world. There will be a new normal at some point, but it will be different from the normal we left in February.

Thank you for this post. I always find your views thoughtful and helpful. I agree with your basic point that the transition to OA is only going to be harder with tight budgets. Rising revenues always lubricates away the pain and inefficiency of transitions – its easier to keep a foot in two worlds when you can afford both of them.

I guess I’m wondering why you see OA transition as such a given. I agree that Plan S has had big effects on continental Europe and even the UK. But I just don’t see anywhere near the same response in North America. One can point to a few cases like U California or MIT. But its not even really a topic of conversation on most campuses I’ve visited. And major funders like NSF have pointedly avoiding jumping on an APC/Gold OA band wagon being content with cheaper, less robust models like green or delayed OA as a way to get research out of paywalls without new costs. Even before the budget crisis most US institutions (and admittedly I mostly hang out with people at net Publish universities) were acutely aware of the costs shifts happening. The increased costs of APC Gold OA were seen as untenable even before new budget crises hit (probably in part because library budgets never even recovered from the 2010 budget crisis).

My crystal ball predicts a world for the next years, maybe the net decade, where Europe and North America are pursuing different models. That’s not good for science or publishers. But I don’t see the enthusiasm let alone resources for a transition to OA in North America.

Thanks Brian, I think these are really good points. I would counter by separating out open access from the APC business model. I think there is indeed widespread enthusiasm for OA (with exceptions around the CC BY licensing terms in the Humanities). The problem is, as you note, that no one wants to pay for it. While there are regions that have largely solved this problem (SciELO and AmeliCA for example), these sorts of systems would be a lot harder to put in place where there is already a well-established industry and system of academic career advancement (for better or worse) that is largely built around it. We need better business models, and while we’ll continue to see progress in regions willing to shoulder additional costs for existing models, we won’t see universal OA, nor a system that is broadly equitable until we find them.

For the record, I do not believe that there is widespread (meaning majority) support for OA. Prestige matters more.

This is a false dichotomy. Prestige and OA are not opposite sides of a coin. I’m willing to bet that if you asked 1,000 researchers if they would prefer that their papers be made freely available immediately for everyone to read or limited to a smaller readership that subscribes to the journal in which they are published, the answer would be near-unanimously (if not entirely unanimous) in favor of the former.

If you required them to sacrifice prestige, or gasp, pay for publication, of course the numbers would be very different. But as I said, if you can separate out the concept of OA from the APC business model and all the problems it creates, it seems an obvious choice, if not an immediately realizable one.

Talk of a false dichotomy! Please don’t become a political pollster, David, or you will be predicting the presidency of Jill Stein. That limited readership you cite is the proper audience for research papers, which are written for other researchers, not for Twitter. In any event, the word “consensus” is a weasel word. It is taken to imply a properly administered protocol and a quantitative outcome. What it means in practice is “friends of mine.” The real measure is not what people say but what they do. Look at the level of submissions. That’s the vote.

Sorry Joe, I think we’re talking past each other here.

I understand the level of paid OA uptake and spend a lot of time analyzing it to make strategic decisions for many journals. Again, that’s irrelevant to the question of whether there is support for the concept of open access. You’re looking at support for one business model, one that I agree is fraught with problems (https://scholarlykitchen.sspnet.org/2019/10/09/roadblocks-to-better-open-access-models/). You’re asking and answering a different question than the one I asked. You’re looking at whether researchers can afford/are willing to pay for open access. I’m asking whether they support/prefer the concept in general, putting aside cost/route.

Can you cite any researcher that wants to limit access to their publications? Do you have any evidence that suggests that authors prefer paywalls and stopping potential readers from seeing their work? It seems an obvious point, and not one worth arguing — what researcher in their right mind would actively prefer to limit the reach of their life’s work? I’m not asking what they can afford nor asking them to set aside career concerns about prestige (although the insane levels of growth for MDPI might make you question that assumption). I’m only asking whether there is widespread support for the concept of open access itself, and I think there is.

And if you merely want to look at what authors “do”, then perhaps worth noting recent data from Unpaywall noting that 53% of articles with a DOI published in 2019 are (at least some form of) open access: https://twitter.com/jasonpriem/status/1283491796732043265

Sorry, David. We must agree to disagree. It’s the way you are framing the questions. Jill Stein for president!

Okay.

But I would reply that you’re stuck in a view that OA equals, and only equals, author-pays Gold APC open access. There’s a lot more going on. I don’t recall hearing a lot of complaints from NIH-funded authors about their papers being made freely available through PubMed Central, nor do I see a lot of complaints from users of ResearchGate that there are free copies of their papers available.

I didn’t realize that stating “people like free stuff” would be so controversial.

Your article really stirred something in me, David. I read it two or three times for two or three days wanting to comment. My thoughts were not positive, and I tried to move on. Then I read another Scholarly Kitchen post that seems relevant and then a LinkedIn post, and here I am trying again.

The article starts with the premise that OA is overall good, a point which has been made more evident by the pandemic. Then it goes on to comment on how the current funding model for OA, the APC, has meant that rich researchers have a leg up on publishing their results OA, getting the benefits of doing so, over their less wealthy colleagues. Later on it describes how the resulting reduction in revenue from the transition to OA has driven independent research societies to turn their journals over to the large commercial publishers. This does not seem to support the original premise.

I fully support your final paragraph. My worry is that when we reach OA nirvana, we will look up and see a world dominated by the rich/wealthy researchers and the large commercial publishers.

Hi Tim, like the conversation with Joe Esposito above, I think the disconnect you’re seeing is in the assumption that open access automatically means author-pays-APC-Gold-OA, rather than looking more broadly at the concept and considering other routes to success. I think you have accurately described the likely destination for scholarly publishing if we continue down the path of APC-driven OA, and there are certainly commercial organizations who are exerting a lot of pressure to drive things in this direction. I’m on the record for years pointing out the flaws of this model (e.g., https://scholarlykitchen.sspnet.org/2019/10/09/roadblocks-to-better-open-access-models/)

But that’s not the only way to achieve the goals of open access. Certainly routes like the US Government’s public access policies, SciELO, AmeliCA, Diamond OA journals like eLife, Subscribe To Open models, community-driven efforts like the Open Library of the Humanities, and scores of others including new models not yet discovered deserve consideration. This is where the leading minds in the move toward open access and open research are all heading, away from an APC model that creates as many problems as it solves.

It’s not a problem with an easy answer, and we are not yet close to solving it. I do think a lot of the market is going to move toward this flawed model because people are impatient, and for some researchers in some fields in some geographies it does at least solve one problem. But increasingly people are seeing the flaws and looking to find other routes.

Since you reference me, I want to state that I made no connection between OA and APCs. My issue is with how surveys are conducted.

Indeed, which is why the point I was making remains opaque to you. 😉

David: Talking ( and more importantly implementing) Open Access, without identifying the financial model that makes it possible, seems somewhere between fantasy and duplicity. Gold, green, Diamond or Psychedelic, I don’t care: Peer review, Selection/curation, audience engagement, and distribution entail real costs. Those need to be covered from somewhere; to talk about OA without identifying where/how it is funded, is as specious as those Plan S robots blindly repeating the “big lie” that there is ‘enough money in the system”. And BTW, as you should know, redistribution, licit or illicit, as in your PMC and Research Gate examples, is not publishing, just amplification, even though that too requires financial backing from some pretty large bankrolls.

I don’t think it’s duplicitous to talk about an idea even though you haven’t quite figured out a way to make it work. Should we stop talking about ending world hunger or global warming because we don’t have an immediate and perfect solution for either? Dialogue is a good way to learn and generate new ideas and new approaches. Experimentation (actually implementing) new models is necessary to see if proposed models work. So far, the models in play do work in some areas of research and in some geographies and at some levels of scale. I’m with you though in that they’re not universal and many models cause as many problems as they solve. But sticking one’s head in the sand is not the way to solve problems and drive progress.

Thanks DC for your perspective. Not attempting arguing for either an end to innovation or exploration in Open Access. Perhaps my use of “talking” was overly slangy. However, given where we are 10+ years from Budapest, where Open Access is being imposed (in select markets), on little more than blind faith and attempted researcher behaviour modification by fiat, I would hope that experimentation would need to disclose both proof of concept and methodology – especially before closing off previously successful methods of publication.

Comments are closed.