By Kevin Smith

Recently there has been a spate of comment expressing frustration about the allegedly slow progress of open access, and especially Green open access. It is hard to disagree with some of this sentiment, but it is important that frustration not lead us into trying to solve a problem with a worse solution.   The key, I believe, to making real advances in open access is to walk away from the commercial publishers who have dominated the market for scholarship. Only if we do that can libraries free up money from our collection budgets to do truly new things. A new business model with the same old players, even if it were possible, would be a mistake.

The most articulate call for an open access future for scholarship – the Budapest Open Access Initiative — was issued fifteen years ago, in February 2002. There is no one better qualified to speak about the meaning of that declaration today than Professor Jean-Claude Guédon, who signed the original Budapest statement and last month published a brilliant and compelling article about where we are and where we need to go next in the movement toward open access scholarship.

Guédon covers a lot of ground in his article “Open Access: Toward the Internet of the Mind.” I want to focus on two points, one I think of as a warning and the other a signpost.

As part of recounting the development of open access from the BOAI to where we stand today, Guédon talks about the development of the APC funding model, where authors, institutions, or funders pay per article charges to make journal articles OA. He notes, and although it is an obvious point it still struck me when I read it clearly articulated, that APCs developed as a way for commercial publishers to exploit the movement toward OA for their own profit. That is, this business model is only a sensible way to finance open access if one accepts that the traditional commercial players must remain part of the equation in an open access world. Not only is that assumption not obvious, it also has some unfortunate consequences.

As Guédon points out, so-called “predatory” journals (Guédon calls them “pseudo-journals”) can only exist because of the APC business model:

Such journals have exploited the APC-Gold business model. One could add that without APCs, the “predatory” journals could not exist. Part of the attraction of the APC model is that it provides a risk-free way to publish, since all costs are covered upfront… The first consequence of this illegitimate business strategy is a pollution of the scientific archive. The second result is more indirect: the presence of “predatory” journals makes all lesser-known titles more questionable (p.12).

This last point is especially important, I think. The APC business model does more than just provide a way for traditional commercial publishers to make money from open access, it also helps to delegitimize perfectly respectable forms of OA publishing because of the taint of “predatoriness.” There are, of course, predatory practices throughout the publishing industry, and they take a lot of different forms. But the attention given to “predatory OA,” which is the bastard child of the efforts by Springer and Elsevier to profit from OA, causes suspicion to fall on any journal that does not have the right branding.

In this context, it is easier to see why the proposal that came out of the Max Plank Institute to “flip” all subscription journals to APC-Gold open access is such a bad idea. It is the fruit, in my opinion, of frustration rather than careful reflection. Major commercial publishers like Elsevier and the STM Publishing association have already rejected the idea, and since their cooperation is key, that in itself should kill the plan.[1] But there is a deeper harm that would result from an all-APC approach to OA.

At the recent ACRL Conference in Baltimore, I attended a session about predatory publishing, which unfortunately focused exclusively on predatory OA practices, but did acknowledge that the root problem, APCs, was a creation of the commercial publishing sector. The interesting data for me, however, were expressed in a map showing that the “victims” of predatory OA journals were overwhelmingly from the Global South. That is, the high cost of “real” APCs is driving scholars from the developing world into the arms of those offering low APCs and little or no actual publishing support.

In a “flipped” OA world, APCs would have to rise further; this is a point made clearly by Elsevier and the STM publishers in their response to the Max Plank proposal. And that would make this problem for the developing world even worse. The “acceptable” outlets for scholarship would be even more closed to those scholars than they are now, and the lure of predatory publishing would be much greater. A proposal to flip scholarly publishing is actually a form of scholarly colonialism, shutting out the Global South, forcing scholars from half the world to chose between predatory publishing or the charitable whims of large Western conglomerates.

If a flipped model that depends on APCs is not the way forward, what is? Here Guédon gives us some clear signposts pointing forward. He calls on libraries to reimagine themselves “as interconnected nodes that help a knowledge infrastructure emerge (p14). ” He directs our attention to the OpenAIRE project in Europe, which, as he says:

While starting from a large set of repositories… has decisively moved to the level of a full network. This leads to the creation of a distributed server system making possible the recovery of any document deposited anywhere in the network. The network is built on top of national systems and it is woven together through a set of protocols and standards to ensure interoperability (p. 16).

Guédon goes on to outline some of the opportunities that such a network built on top of repositories make possible, including linking publications and their underlying datasets, open peer review, and text and data mining.

In the U.S. we do not have the advantage of national networks, of course, nor the investment of centralized funding, for the most part. But we do have a resource that might offer a starting point for this next step in making open access a more robust and sustainable approach to knowledge creation and sharing – the consortia to which many of our libraries belong.

Our consortia often function as primarily as “buying clubs” that negotiate to reduce the cost of purchasing commercial products. The future of that role, however, is doubtful; more and more libraries are backing away from “big deals,” either because they have to for financial reasons or because they realize that these deals, while good for commercial publishers, and not in the best interest of our universities. Consortia need to redefine their roles every bit as much as libraries do, as the movement away from commercial publishers accelerates.

I want to suggest three ways that library consortia can help universities begin to wean themselves from the unsustainable system of scholarly communication that developed in the era of printing and begin to usher in the next stage of open access scholarship:

  • Consortia are well placed to develop the infrastructure to connect our institutional repositories and to develop services on top of them. This step would increase the effectiveness of Green OA exponentially, and open up opportunities, such as text and data mining, that we currently seem to rely (wrongly, imo) on permission from commercial interests to exploit.
  • Consortia could help broker a coordinated approach to library publishing, where specific groups of universities take on particular disciplines in which to publish. Thus strong nodes of disciplinary expertise and a critical mass of materials could be created. Such coordinated publishing, undertaken in collaboration with university presses and scholarly societies, could be a key to breaking the grip that legacy publishers have over scholarship.
  • Consortia could organize conversations amongst provosts and presidents about how scholars and scholarship are evaluated. Libraries often think of this as an intractable problem, and political forces, as well as inertia beset our conversations on our own campuses. One way to break this logjam is to gather together a critical mass of campus leaders and guide them in a conversation with each other. For the most part they know that the current system is absurd, using surrogates for quality that were never intended for that purpose. In fact, we spend millions – billions, collectively – to purchase social capital from commercial entities, even while the work that builds that capital is done pro bono by our own faculty. More than anything else, this system cries out to be reclaimed by our universities. Consortial conversations, where our campus leaders have a chance to talk and bargain with each other, might be the way forward.



[1] I have read Elsevier’s “Thoughts on the Max Plank White Paper” and have a copy in my possession. I have been told, however, that it is a private document, and I was asked not to share it widely. The arguments it makes, about why publishers are unwilling to give up the subscription model and how APCs would have to rise in any “flipping,” are essentially the same as those expressed in the STM statement.


Kevin Smith

Kevin Smith is a librarian, a lawyer focusing on copyright issues, a scholarly communications advocate, and the Dean of Libraries at the University of Kansas.

Comments (19)

  1. He notes, and although it is an obvious point it still struck me when I read it clearly articulated, that APCs developed as a way for commercial publishers to exploit the movement toward OA for their own profit.

    Really? I thought APCs were first introduced by the born-OA publisher BMC and then adopted by the non-profit PLOS. Unless you know different, I thought that the legacy publishers’ adoption of APCs was rather a Johnny-come-lately manoeuvre.

    1. That’s true, Mike, and it’s also true that vanity presses have been around for forever, way before Open Access, so the whole predatory OA thing is just a sideshow. I don’t know how one can argue that poorer countries will suffer under Gold OA without considering whether the rich countries will subsidize the APCs of the poorer ones, which is exactly how it works at PLOS.

  2. Guédon provides a good overview of the development of open access, but it suffers, as many such overviews do, of focusing almost exclusively on STEM, virtually ignoring HSS, which is not always well served by the OA approaches first developed for STEM fields. That point aside, though, the larger issue of how to wrest control away from commercial publishers and reverse the priority of business models over communication needs, which is the main theme of Guedon’s piece, does deserve more attention. Kevin points toward library consortia as a possible focal point for such an effort, and that makes sense. But such consortia themselves are not well equipped to carry out the function of evaluation that Guedon highlights alongside communication as the two necessary elements of the system. Kevin only in passing refers to university presses and societies, but it is those bodies that actually have the trained staff and skills that are needed to perform the evaluation function adequately. Libraries, even those that have started publishing operations, still lack this skill set. So it seems to me that university presses and societies must be fully integrated into any scheme that has a chance of challenging the dominance of commercial publishers. University presses were doing this business long before commercial publishers entered the scene, and they should, with proper support, be able to take this territory back under university control.

  3. Kevin, This sentence propagates a myth that is simply not accurate:

    “[Jean-Claude Guedon] notes, and although it is an obvious point it still struck me when I read it clearly articulated, that APCs developed as a way for commercial publishers to exploit the movement toward OA for their own profit.”

    In fact, as Jean-Claude himself notes, APCs were developed virtually contemporaneously by PLoS and BioMed Central – the first, a non-profit publisher founded by scientists for scientists with a wholly transformative purpose, and the second, a commercial but decidedly non-traditional publisher that was once similarly a darling of the OA movement. Nucleic Acids Research, published by non-profit academic publisher OUP, adopted author charges in conjunction with a subscription-membership arrangement soon thereafter, and the American Physiological Society was experimenting with a similar strategy around the same time (as others probably were as well). So let’s not resort to revisionist history in the rush to associate APCs with the taint of commercial publishing. While it’s true that Springer adopted its hybrid APC in 2004, most other commercial publishers have implemented APCs only reluctantly in response to pressure from funders and policy makers.

    There is much more to say about the ideas in your post, but I’ll limit myself to two points here. First, although the problem of predatory publishing is real, there is quite a lot of debate at the moment about what constitutes predatory as opposed to low-cost publishing – Beall’s list hasn’t always been helpful in distinguishing the two. Second, the notion that the OA2020 movement launched by the Max Planck Society is designed to simply maintain the publishing status quo via APCs is, in my view, a misreading of the goals of that initiative. In fact, I think your recommendation to simply cancel all big publisher deals is perfectly consistent with OA2020, since its basic premise is redirection of library budgets away from subscriptions and toward open access. Go for it! Whatever one thinks of OA2020 (and views at the University of California are decidedly mixed), its ultimate goal is to open up publishing (and library budgets) to further transformation, not suppress it. Many if not most institutions that support OA2020 are equally committed to supporting the many non-APC models that are taking hold, particularly in non-STEM fields (Online Library of the Humanities is a great example). These are not either/or propositions, but complementary efforts aimed at largescale transformation in the service of open scholarship. Whether one does or doesn’t agree with the argument that an APC model will drive competition, APCs may actually incentivize this transformation, by exposing the economics of the current system in a vivid and impactful way.

    Ultimately, the future face of scholarly communication will be determined by the scholars and scientists themselves, along with their societies and the academy as a whole. If the new scholarly communication infrastructure that you, Jean-Claude and others espouse is embraced by scholars, institutional support will inevitably flow in that direction.

  4. Kevin, while I agree with a lot of what you say here, this statement, attributed to Jean-Claude Guédon, is manifestly incorrect:

    “He notes, and although it is an obvious point it still struck me when I read it clearly articulated, that APCs developed as a way for commercial publishers to exploit the movement toward OA for their own profit. ”

    There were various experiments with APC or APC-like models in the 1990s, but “gold” open access publishing took off in the early 2000s with its adoption by pioneering open access publishers BioMed Central and PLOS. The use of APCs was a logical outgrowth of the reconceptualization of scholarly publishing as a service to the community, as APCs allowed for journals to recoup their costs while making their contents immediately freely available. Far from being a means for commercial publishers to exploit the movement to OA, APCs were necessary to fund the establishment of OA as a viable publishing model. At the time the APC model was widely derided by established players in the industry – large for profits like Elsevier, Wiley, Springer, Nature and virtually all non-profit society journals. It was not until BMC and PLOS achieved commercial success that the model became more widely adopted by large commercial entities.

    1. I should add for clarity that I am one of the founders of PLOS.

  5. I apologize for the careless language; the error is entirely mine and should not be imputed to Professor Guédon, who clearly knows better. Still, I am not sure that the point makes any real difference. APCs remain the business model that is allowing the commercial publishing sector to co-opt open access to their own profit and control, and proceeding with any plan to flip scholarly publishing to OA using APCs is still a very bad idea.

    1. Kevin, I want t to take up’ll just take up the myth you are perpetuating that “APCs remain the business model that is allowing the commercial publishing sector to co-opt open access to their own profit and control”. This fear that many express is simply contrary to the basic economics of market power.

      Traditional publishers have market power plain and simply because *authors* want sufficiently strongly to publish in traditional journals with reputation (often measured, poorly but widely, by impact factor) that they are willing to give up their copyright to do so. The only sources of market power (the power to charge prices above cost, thus earning monopoly profits) for journals are this copyright ownership over their published content, and the reputation they have developed based on the history of copyrighted content they have published. Flipping to OA (paid by APCs, or through other business models) eliminates the sole underlying source of market power (copyright over published content). It does not in any way protect or increase the journals market power. The only remaining source of market power is reputation based on past publications, but reputation for (all) products can and does change over time. How fast new journals or other forms of scholarly communication might rise in an APC world is much debated, but it will be no *harder* for them to rise than it is in the current subscription-based world.

      APCs do *not* increase publisher control. Indeed, as you note, many major publishers (like Elsevier) are strongly resisting full OA (or hybrid with full offsets against subscription payments); they would embrace it if it would increase their profits and control. As several commenters have noted, APCs originated with pure OA journals, and traditional publishers generally have been dragged along kicking and screaming.

      Indeed, APCs will *reduce* publisher control, if paid in a way that aligns incentives between authors and libraries (or others paying subscriptions). In one version of the APC world, libraries simply use their subscription budgets to pay APCs on behalf of authors. Then the power balance is unchanged: publishers and libraries negotiate over the total payments a university etc. will provide to the publisher. But if authors are responsible for the payments (at least in significant part, but perhaps shared with their library or research office), then they — and especially important, the editorial boards composed of authors, which are essential to the success of traditional journals — have a reason to at least partially care about the cost of where they are publishing, and journals will have a reason to compete with lower APCs because unless they attract content, they have nothing. (Yes, authors care about reputation / impact more than cost, but if librarians would spend more time talking to authors they’d know that for nearly all articles scholars write, there is more than one acceptable place in which to publish it, and no one submits all their articles just to “the” single best journal in their field. And all that is necessary for competition is that there be two or three journals that most authors will contemplate much of the time for many of their articles. And we know that authors make economic tradeoffs like this all the time in decisions about hiring research assistants, post-docs, purchasing equipment, deciding how many and which conferences to travel to, etc.)

      So, APCs create *no* new control, and they provide the possibility of substantially decreasing it, by empowering authors, who have the ultimate control: over the decision about where to submit their articles, and thus to whom (or whether) to grant copyright.

      Of course, as Ivy correctly notes in a comment above, APCs are not the only way to pay for open access journals. Indeed, recently we’ve seen some exciting examples of funding providers directing funds to publishers so they will publish OA: e.g., the Gates Foundation with Science (, and the Robert Wood Johnson Foundation with Annual Review of Public Health ( The University of California is providing funding to enable low-APC OA publishing by the UC Press for journals and monographs (Collar and Luminos,

      Publishing — whether by traditional publishers, or societies, or University presses, etc., etc. — costs money. We have to get sufficient money into their hands for the value added that authors and libraries and readers *want*. Though some prefer various approaches over others, the OA2020 Expression of Interest merely asserts an intent to redirect subscription monies to alternative funding models that support OA publishing.

      1. I largely agree with this response, but it raises an interesting question. Since a power imbalance exists between individual authors and giant publishers, what is universities were to exercise their theoretical right, under copyright law, to consider faculty (academic) publications as “work made for hire,” which would then make the university, not the individual author, the bargaining agent with the commercial publishers. They would presumably have much greater leverage to negotiate copyright terms than isolated individual faculty members do. Could this be a viable model for higher education to take back control of scholarly publishing?

        1. Sandy, I don’t the issue here is about article negotiations over specific copyright terms. There are two big issues (at least in this conversation, I think): is published material open access as a standard matter (not article-by-article)? And how much money gets paid by society to publishers?

          Various pre-publication payment models can lead to open access publishing (consortial agreements like SCOAP3, funder payments like Gates and Robert Wood Johnson, university publishing, APCs, etc.). The concern that Kevin and others raise is whether, if we turn to such business models — Kevin focuses on APCs — publishers will have more power *over price*. They already extract profits from us through subscription charges. Will they be able to extract more profits from us through APCs? My point is that switching to pre-publication payment funding of publishers (whether APCs or any of the other models) does *not* increase their market (price) power, and in fact, if done well, *reduces* publisher power. And here is where the myth that the poor, individual author is small and weak compared to the big nasty publisher is wrong: the author has *complete* power over where to submit. There is always at least some choice between journals of comparable prestige: if Nature is charging the highest price, submit to Science or Proceedings of the National Academy of Science or Cell, etc. The publisher has *nothing* if it can’t attract good articles, and the authors have complete control over that.

          1. Do individual authors really have the kind of power you attribute to them? Especially for tenure-track junior faculty, their choice is constricted by what their senior faculty colleagues think about which journals and publishers are the most prestigious. I can cite any number of instances when I was director of Penn State Press and had persuaded a junior faculty member that ours was the best publisher for his or her book, yet that individual had to go with the publisher that the department head considered most prestigious. Senior faculty may have more freedom in this regard, but even they have to pay attention to which journals and book publishers their colleagues esteem the most. And it is this prestige, of course, that is the ultimate source of publishers’ power, even though it only exists because of what faculty think.

          2. Thanks for the discussion on this, Sandy. Your concerns are an example of the lost nuance that requires more direct input from faculty.

            Yes, junior faculty need to publish in prestigious journals; mid-career faculty too (they want to be promoted to full), and even if senior can “get away” with less prestigious journals, they generally don’t want to — they want increasing prestige, and more prestigious journals generally get read more, and every author wants to be read!

            However, the myth that many librarians discussing open access keep repeating is that this leaves scholars with “no choice”: they have to publish in “the” prestige journal. But that’s wrong. Every author writes articles on various topics (even if they specialize to a problem, there are different aspects of that problem), and writes articles of different quality (*no one* publishes *only* in, say “Nature”). And for every article, even the very best, there are always, in every field, more than one journal of appropriate quality for the article. At the very top for STEM fields, for example, there are Nature, Science, Cell, Proceedings of the National Academy of Science, at least. Publication in any of those is considered “star quality”. For articles not headed to that tier, there will be multiple journals of comparable quality. For example, in economics, there are four “general coverage” journals that are considered the premier level: American Economic Review, Journal of Political Economy, Quarterly Economic Review and Econometrica. Publication in any of those confers approximately equal prestige. For lesser articles in the subarea of, say, industrial organization, there are another 4 journals or so that are “second tier” but all roughly equal in prestige.

            In other words, prestige yes, but there is almost never such a thing as *the* “most prestigious” for any article (or book): there are always at least 2-3 choices (usually more), and that’s enough to get genuine price competition.

            (FWIW I say this based on substantial experience: I’ve been a tenured professor at U Michigan and UC Berkeley for 31 years now; I’ve been on lots and lots of tenure committees and written many outside tenure review letters for scholars at top universities; and I’ve been first an associate dean for academic affairs (in charge of the promotion and tenure process), and then the dean of a school at Michigan (responsible for final tenure and promotion recommendations to the provost). On my own I’ve published over 80 peer-reviewed articles and for *every one*, my coauthors and I always had a discussion considering to which of several comparable prestige journals to submit, considering things like closeness of editorial match (never completely obvious), typical turnaround time, who is on the editorial board, etc. In other words, after we settled on a prestige “class” of several comparable journals, we’d consider many other factors. If price were a factor, we would have considered that too.)

          3. All points well taken. But I can tell you this much. When i was en editor at Princeton University Press (1967-1989), i could compete for books on equal footing with Cambridge University Press. At Penn State Press (1989-2009), even though everything else was the same as at Princeton (quality of copyediting and book design, high manufacturing standards, same rigor of peer review, etc.) I could seldom compete successfully with Cambridge for books by junior faculty and eventually gave up trying, even though Cambridge’s books were generally priced much higher than ours were.

            Also, as far as my own writings are concerned (80+ articles too), since I was on no temure track and publications had no effect on my career advancement, I had tiotal freedom about where to publish and the journal’s OA policy was a prime consideration. All of my articles, now posted at Penn State’s IR, are freely accessible, though I prefer the CC BY-NC-ND license (which is the default license at the IR) to the CC BY license favored by the BOAI crowd.

  6. […] she winced all the harder to read his encomium to consortial action on open access. The Loon is sorrier than she can fitly express to say this, but that post is more wrong than it is […]

  7. Kevin, I’d like to tackle some of the next steps you outlined, to clarify our thinking about how to best use increasingly scarce resources from libraries to advance OA.

    First, you propose to “develop the infrastructure to connect our institutional repositories and to develop services on top of them.” This, of course, is what initiatives like OpenAIRE and BASE in Europe, and SHARE in the U.S., are doing, building on a long history going back to OAIster and OCLC. And let’s not forget Google Scholar, Academic Search, Meta, and all the other commercial products that aggregate and add value to OA content in institutional and discipline repositories. To the extent that aggregation and added-value services can be built, we’re already doing it, and it’s having very little affect on publishing practices. It also costs a lot of money on top of what we still need to spend on subscriptions.

    Second, you propose to “broker a coordinated approach to library publishing, where specific groups of universities take on particular disciplines in which to publish” in coordination with university presses and societies. Presumably this means creating a shadow system of approximately 30,000 current scholarly journals (commercial and pseudo non-profit), then convincing nearly 2M academic authors to switch to them, with no particular incentives and several downsides (e.g. lack of editors and peer reviewers, lack of reputation, etc.) If we’re worried about the cost of flipping existing journals to OA business models, this sounds even more expensive.

    Third, you propose to “organize conversations amongst provosts and presidents about how scholars and scholarship are evaluated.” This is absolutely at the heart of the problem and I wish it were a question of convincing university administrators to change their ways, but the reality is that our system of academic recognition and reward is perpetuated by exactly the same academic authors whose behavior we would like to change. Maybe my institution is an extreme case but it such a change to academic recognition could never be achieved without the strong support of faculty who are still very invested in the status quo. There might be other ways to spark those conversations, via societies for example, and funders definitely have influence, but that’s going to be a long and difficult process.

    We need to think more carefully about where libraries have real influence and where we should invest to advanced hange, or we’ll find ourselves unable to sustain the current journals *or* fund alternatives…

    1. OA2020 has always been a transitional strategy, one that recognizes the impracticability, as MacKenzie notes, of wholesale reconstitution of a very large portion of the scholarly publishing system. It is not simply “the fruit…of frustration,” although the lack of progress is frustrating to everyone, but the result of the need for a plausibly doable transition strategy to get from where we are now to both more OA and a competitive OA publishing market (supported by a diversity of business models, not just APCs).

      The fact that the the major commercial players are not yet willing to seriously entertain offset agreements is already clear. Why should they? Subscription revenues are robust and they’re benefitting from the new hybrid and Gold OA markets. I agree with Kevin that “the key… to making real advances in open access is to walk away from the commercial publishers who have dominated the market for scholarship.” Until subscription revenue goes down, subscription publishers clearly have no incentive to move from where they are now; but when titles receive many cancellations, the OA business model will begin to look more attractive flipping subscription titles to Gold OA will (eventually) look financially appealing. It may not take widespread cancellations, either. The publishers’ strong desire to lock customers into multi-year Big Deals version 2 based on current prices (an offset deal) is likely to bring them to the table before the point where profit calculations favor Gold OA across their entire portfolio.

  8. […] recently, Kevin Smith furthered this conflation when outlining his skepticism of APC-funded open access. In that piece, Smith  identifies the predatory publishing problem as at the heart of […]

  9. […] began to propose the outlines of a long-term strategy in this blog post from last month.  I was interested to see that a short-term strategy was […]

  10. Thank you, Kevin, for sharing your views! In your last bullet (Consortia could organize conversations amongst provosts and presidents…), you end by saying, “Consortial conversations, where our campus leaders have a chance to talk and bargain with each other, might be the way forward.” Can you please describe this a bit more? Do you mean campus leaders will bargain over how, “specific groups of universities take on particular disciplines?” What other issues do you think campus leaders will bargain over?

Leave a Reply to Consortial institutional repositories and the bystander effect – Gavia Libraria Cancel reply

Your email address will not be published. Required fields are marked *