As anyone who has followed the story of open access will know, a multitude of issues has arisen since the 2002 Budapest Open Access Initiative (BOAI) adopted the term in order to promote the idea of research being made freely available on the internet. It has also led to a great deal of debate and disagreement over the best way of making open access a reality.
However, we seem to be arriving at the point where consensus is growing in the global North around the idea of persuading and/or forcing legacy publishers to convert (“flip”) all their journals from a subscription model to an open access model.
One implication of this would seem to be that we can expect widespread use of the pay-to-publish model where, instead of readers paying to access other researchers’ papers, authors will pay to publish their own papers – by means of article-processing charges (APCs). Currently, APCs are around $3,000 a paper, although they can be both higher and lower than this.
OA Big Deals
The argument for a global flip has been most fully articulated in a report published by the Max Planck Digital Library (MPDL) in April 2015. This concluded that “An internationally concerted shifting of subscription budgets [to open access] is possible at no financial risk, maybe even at lower overall costs.”
Using this report as a foundation, in 2016 MDPL established the OA2020 Initiative, which was launched at the (controversial) 12th Berlin Open Access conference. The stated aim of OA2020 is to “convert the majority of today’s scholarly journals from subscription to Open Access (OA) publishing.”
The MDPL report is also driving the current trend for OA Big Deals. These are being negotiated in Europe (usually by national consortia) with legacy publishers. The aim is to facilitate a transition from a subscription-based world to a pay-to-publish world, by means of agreements that combine subscription payment for paywalled content with a bulk payment to provide OA publishing rights for researchers to publish their papers OA without themselves having to find the money to pay APCs.
This model was pioneered by the Association of Universities in the Netherlands (VSNU), which is one of the OA2020 signatories.
Meanwhile, following the publication of a report from the University of California, there is a similar trend emerging in North America. The UC report concluded that a move to an APC model “could be successful over time, following a necessarily complex transition period.” The University of California is also a signatory to OA2020.
To encourage other US universities to go down the flipping road, in March this year the University of California launched its Pathways to OA initiative. Among other things, this proposes emulating the European OA Big Deal model (or offsetting agreement, as it is alternatively called). As a group of University of California OA advocates explained recently in Nature, “UC libraries will explore negotiating offsetting agreements to drive the transition of hybrid journals to becoming fully open access. The strategy involves setting transformation benchmarks and then, during the transition period, offsetting an institution’s spending on open-access article processing charges against the total price of its subscription package.”
OA Big Deals are nevertheless controversial, not least because there are concerns that they will consolidate the malevolent hold that some believe legacy publishers currently have over scholarly publishing.
Certainly, researchers in the global South view a mass flipping of subscription journals to OA with considerable concern. Since most have little or no access to APC funding (and are extremely unlikely to benefit from the hugely expensive OA Big Deals) they can expect to see today’s paywalls replaced by publication walls, making it extremely difficult for them to publish in international journals.
Alternatively, in response to European demands for flipping Elsevier has mooted what it calls “region-specific” open access. This would see European articles made available as immediate gold open access within Europe, but restricted to “green open access outside of Europe”.
Two points of view
One of those concerned about the implications that these developments could have for the developing world is Jamila Jaber, Library Director at the Islamic University of Lebanon (IUL). Last year she contacted me to ask if I would consider doing an interview “with two researchers, one from a ‘developed country’ and one from a ‘developing country’ (from the Arab world for example).”
The aim, she explained, should be to allow for a discussion about scholarly communication and open access from two geographically different points of view.
It seemed like a good idea, so I asked Jaber if she could propose a candidate from the Arab world, and she suggested Mahmoud Khalifa, who works as a librarian at the Library of Congress Cairo Office and is DOAJ Ambassador for the Middle East and Persian Gulf. Khalifa is also President of Cybrarians, which publishes an information science journal and runs two conferences.
To provide a voice from the developed world I invited Jeffrey MacKie-Mason to take part. MacKie-Mason is UC Berkeley’s University Librarian and Chief Digital Scholarship Officer. He is also an enthusiastic advocate for journal flipping. MacKie-Mason argues that engineering a mass conversion of subscription journals to OA is currently the only practical way of achieving open access in the near term, and that while a global flip presents challenges for those in the global South, the current paywall situation for them is “awful”. He adds that we cannot expect open access “to remedy all inequities”.
Elsewhere, we could note, Leslie Chan, has argued that “The institutions and countries adopting the OA2020 initiative express very clearly that it is not their problem that scientists from developing countries can publish or not. It is a very selfish attitude, individualistic and even nationalistic.”
The challenge for me in organising a discussion between two people from, respectively, the North and the South was how to combine input from the two interviewees in a formal interview process. After thinking through various possibilities, I decided to do two separate interviews, and then invite each interviewee to comment on the answers provided by the other interviewee.
It is not an ideal approach, but it might at least help focus (much needed) attention on the South/North question and give us a sense of how open access (and strategies for achieving it) can look very different in different parts of the world.
I would hope that others will comment too, since if these issues are not discussed fully now those in the global South could find that OA has made things worse rather than better, and that they have become locked out of the global research conversation in an even more insidious way than they are with the subscription system.
I think we are also invited to ask whether a global OA solution is actually possible. If it is not possible, then we might wonder how the BOAI’s promise that OA would enable the world to “share the learning of the rich with the poor and the poor with the rich … and lay the foundation for uniting humanity in a common intellectual conversation and quest for knowledge” can hope to be realised.
Below I publish the first interview, with MacKie-Mason. As well as being University Librarian at Berkley, he is Professor of the School of Information and Professor of Economics. He was formerly the Dean of the School of Information at the University of Michigan.
MacKie-Mason has been a pioneering scholar in the economics of the Internet, online behaviour, and digital information and has authored more than 85 publications in economics, computer science, law, public policy, and library science journals.
Note: I realise my reference to The Lord of the Rings in the penultimate question below may be a little obscure for a global audience. And as Jaber pointed out to me, Tolkien saw the “one ring” as a malevolent force. However, I think there was some ambiguity in his portrayal of the ring, and I want to make the point that any force unleashed by trying to manufacture a global solution for open access could prove malevolent or benevolent, depending on its impact on the entire research community (an impact that we can only really guess at right now).
I think it behoves us, therefore, to consider carefully what those implications might be before rushing ahead. And to appreciate that, unless a global solution has the interests of all researchers in mind, it cannot hope to fully realise the vision articulated by BOAI.
The Q&A begins …
Q: Can you say something about yourself, your institution, and why and when you started to take an interest in open access?
J M-M: I am currently the University Librarian for UC Berkeley, but I have a somewhat unusual background. I have a PhD in economics, and was a professor of economics and public policy at the University of Michigan for 29 years. I came to Berkeley as Librarian two years ago; I’m also a professor of information and economics.
Since the early 1990s, my research has been on the economics of information production, dissemination and use, and the behavior of individuals in online environments. I’ve been studying access to scholarly communications for about 20 years; for example, I was the research director for the PEAK project, which experimented with different pricing models for access to 1200 Elsevier online journals, before Elsevier had released ScienceDirect.
I’ve been a strong open access advocate for at least 25 years; all of my research outputs have been available open access at the Michigan institutional repository since the 1990s.
Q: How would you characterise the current state of open access in a) your institution; b) your country; c) your region?
J M-M: Berkeley is very active in open access. According to a recent study by Lokman Meho at the American University of Beirut, we have the 8th highest percentage among US universities of articles published open access. Several of our faculty have pioneered open access journals (including PLOS).
However, generally speaking – at Berkeley and throughout the US – open access publishing is still quite low: we are at 11.2%, and the highest percent at a U.S. research institution is 13.4% (2012–17).
More articles are available than this in pre-publication form through various institutional and subject-matter repositories, but the highest percent available (at least, from legal, copyright-respecting repositories) is about 50%, and most institutions report something closer to 15%.
Q: What do you consider to be the main obstacles and challenges facing the open access movement in your country/region right now?
J M-M: What we’ve seen is a lot of enthusiasm among librarians and institutional leaders, but most faculty at top research institutions appear to be somewhat indifferent, at least as far as taking action.
This is understandable: for promotion and tenure, faculty pay most attention to the prestige of the journal to which they submit, not whether it is open access. Though there is a growing body of evidence that open access articles are more widely read and cited, that’s recent news, and in any case most faculty at top research institutions seem to believe that other top scientists – the ones they care most about reaching and impressing – have access.
It’s not that faculty don’t care about the rest of the world, it’s that they don’t have much incentive to counterbalance the strong career pressure to publish in the most prestigious journals (which for the most part are not open access).
To see widespread (we hope someday universal!) open access publishing amongst US research universities, we’re going to need to either see the existing prestige publishers convert fully to open access (not the bastard child known as “hybrid OA”), or for new, open access journals to gain enough prestige to attract articles away from the current prestige journals – but that could take decades (if it ever happens). We’ve seen new open access journals enter and compete for prestige for 25 years now, and they haven’t made a dent in the flow of manuscripts to the incumbent prestige publishers – in fact, the share of articles published by the top 4-6 publishers has nearly doubled over the past four decades!
Q: What do you think the priorities ought to be in order to move OA forward in your country/region?
J M-M: I’ve argued in print and in many invited talks that given the deeply entrenched – and functional! – promotion and tenure (P&T) system in research universities, we will not make significant medium-term progress on open access unless we find a way to move to a system in which the incumbent publishers switch to full open access.
There is simply no way to drastically change the P&T system across the US (much less the world) in the foreseeable future – and it’s not clear we should (it has evolved over many decades, and all in all works pretty well). Concern for OA is not important enough to transfer the core authority, validation and certification system unpinning science!
This is a relatively unpopular view among US institutions: most despise the exploitation in which the large prestige publishers engage, and want nothing to do with changing the publishing system in a way that lets the incumbents stay in business.
I’d be happy to see the incumbents disappear, too, but I have yet – after some 25 years engaged in study and debate about open access – to see a feasible path forward that doesn’t include the major publishers.
So, for me the priority is to try to convince more and more of my colleagues to coalesce around a feasible solution, and then to jointly use our leverage to force the publishers to change to open access.
The Europeans are making quite significant progress on this front, but to tip the publishers we probably need strong support also from US and / or Chinese institutions. The coordination problem is harder in the US than Europe because our universities are for the most part independent of our national government.
Q: Do you think that open access solutions for the global South need to be different to those implemented in the global North, or should we aim for a global solution? Is it practical or possible to have different geographical approaches to open access?
J M-M: Yes and no. For high-end scientific results, dissemination is and should be worldwide! And of course, the value of open access to society is greatest for the most important scientific results, so we definitely need a solution that works everywhere for publishing this kind of work. I would never work for a solution for the prestige journals that didn’t work for the global South as well as for the global North.
There is a significant amount of research that is of primary relevance to particular regions, and for that regional solutions can be sufficient. And the particular business model supporting open access regional publication can vary from region to region, of course.
Several approaches to open access require shifting the financial burden of scholarly communications from readers (who pay for subscriptions, at least their institutional libraries do on their behalf – but either way it’s a charge for reading) to authors (or their institutions).
By providing publishers with the funds they need to do their work at the time of publication, it is possible for them to do the necessary work of publishing but then make the publications available open access.
Some people express a concern that authors in the global South can’t “afford” to pay for publication, and thus will be excluded from the worldwide scholarly publishing community. I think this is largely a misconception. If we move from a subscription world to a pay-for-publication world, the money currently spent so that scholars and others in the global South can read – the subscription dollars they are currently paying – can be redirected to payment for publication. There’s no reason to think that the total cost will go up (in fact, there are good reasons to think it will go down).
So, at least to first order, it’s a problem of redirecting the flow of funds from one use to the other – but both uses are directly in support of the communications necessary for scientific enterprise, and so it stands to reasons that governments and other funding providers would be willing to put in as much money to publishing charges for open access as they are for subscriptions for closed access. (Indeed, more scholars in the global South will be able to read scientific results in an open access world, so the global South should benefit from this shift.)
Q: You say that given the way P&T systems operate there is no way widespread open access is possible unless/until incumbent journals convert to open access. You also say you believe that P&T systems work pretty well, so would presumably not want to risk subverting them by imposing a radical alternative. I suspect many might disagree that the current validation and certification systems are functional, not least because they tend to rely on metrics like the Impact Factor, which many believe to be deeply flawed when used for these purposes.
Have I understood correctly? If so, how would you respond to those who believe the current P&T systems are not fit for purpose and that clinging to them is actually holding back open access?
J M-M: This is a complicated question, with several aspects.
First, to be clear, I agree that Journal Impact Factors are broken, and have opposed heavy reliance on them for years. More generally, I’m not defending everything about the way that every department, dean, and provost goes about doing promotion and tenure evaluation.
However, the system *is* reasonably functional: it’s a complex set of social norms and practices that have evolved to address a very difficult problem (professional performance evaluation, made even more complex than normal by tenure and thus the need to forecast performance into the indefinite future). As with any complex social system, it could be improved, yes...but doing so is a gnarly problem of great complexity and coordination challenge.
I’m not an expert on performance evaluation systems, but I’ve been a faculty member for 32 years and have seen this system work in depth. I *haven’t* seen coherent proposals for how to reform it in a way that is dramatic enough to largely eliminate the need (in the medium term, foreseeable future) for the prestige publishers. We need a well-developed, labor-intensive peer review system to ensure that our science is valid and useful. And that review system needs some quality markers that people accept (trust).
We haven’t found good ways yet to replace the name-brand reputation of editorial board members, and the accumulated reputation of quality publications accumulated over time by prestige journals, to satisfy peer scientists (I use the word broadly to cover all academic disciplines) about research productivity and quality. (We don’t *just* rely on the quantity of peer-reviewed articles in high reputation journals to make P&T decisions of course – we also solicit external review letters and do some of the reading and evaluation locally with peer review committees – but I’m not aware of any good research universities that consider these other sources of information *sufficient* – they have flaws, too.)
Now, there are some who are sure that they know a “better way”. I’m skeptical – the proposals I’ve seen tend to be naive, rife with magical thinking.
However, even for a better idea, there is also the huge coordination problem: how do we convince the scientists of the world that the new idea is better, and get them (in the medium, foreseeable time frame) to transform their processes that they have developed over decades?
Social transformation of an extraordinarily decentralized system takes more than someone who thinks they have a better idea standing on a soap box. (I’ve seen a number who say we just need to convince a significant number of provosts and chancellors, but that reflects a deep misunderstanding of the primary forces determining P&T: these are faculty-driven processes, that require faculty trust to work, and getting instructions from above to do it differently simply won’t work.)
Much as I *want* us to undercut the power of the prestige publishers – and I do, and have been quite public about this – I don’t see any pragmatic (medium term) path forward to doing so through drastically changing the way we do. P&T. Someday, probably, yes. But I’m too impatient to pin my hopes on this dream.
RP: Am I right to think that you also feel that OA is not important enough to justify changing the current ways of validating and certifying research?
J M-M: Well, that seems a bit like the classic “when did you stop beating your wife?” question.
I think that OA is very important, and it is definitely important enough for us to justify investing substantial time and money to change the way we distribute scholarly communications. And as part of that, I’m delighted that some folks are working on ways to improve how we validate and certify research, either for making those processes more supportive of OA publishing or for other good reasons.
I just haven’t seen proposals that would preserve what I consider to be the absolutely vital role of validation and certification, while eliminating our current reliance on prestige publishers.
If I’m wrong – if there *is* a pragmatic solution that doesn’t throw out the baby with the bathwater, terrific! I’m not wedded to our current methods of validating and certifying, but I *am* wedded to not substantially reducing their effectiveness, and I haven’t seen a better alternative that appears to be feasible.
I simply think that folks who are counting on this approach to solving OA are going to be disappointed, and I don’t want to sit around and wait for this solution that I don’t think is pragmatic in any reasonably foreseeable time frame.
Exchanging paywalls for publication walls?
Q: You say that when people argue that authors in the global South would not be able to afford to pay to publish if there was a global flip to gold they are under a misapprehension because flipping would free up subscription money and this could be redirected to paying APCs.
Is it not the case, however, that most countries in the global South have never been able to afford much in the way of subscriptions, and so have little money in the system? If that is right, surely a mass flipping will for them simply exchange paywalls for publication walls?
Is it perhaps your view that it is not the job of open access to correct inequalities, but simply to make as much research freely available as possible?
J M-M: What I’m saying is basically that the total cost of scholarly communications won’t go up – there is nothing intrinsically more costly (in an economic sense: it doesn’t take more time or other resources) for publishing OA than publishing behind a paywall.
If the amount of money the global North is currently providing to publishers (in the form mostly of subscription payments), plus the amount of money the global South is currently providing (however little that may be) is still transferred to publishers in an OA world (through APCs rather than subscription payments), then there will be more than enough money to support continued publication (I say more than enough because currently the scholarly publishing industry is earning above-competitive profits – that is, more than they need to stay in business).
The question then is just whether there is no practical way to change the flow of funds that makes it no more difficult for scientists in the global South to participate.
I’m assuming that generally, authors don’t publish in journals they cannot read – so I’m assuming that the journals relevant (for the most part) for global South authors are those to which they *do* have access. Their institutions generally do subscribe to some journals. It’s my understanding that in some cases publishers provide steep discounts to enable them to *read* (subscribe). Then, in the gold OA world, wouldn’t we expect those very same publishers to provide steep discounts to authors to enable them to *publish*? The publishers don’t want to exclude the global South from the scholarly communication ecosystem.
Whatever combination of methods are currently in use (discounts from publishers, subsidies from foreign aid to global South research institutions), etc., presumably will be there to ensure at least as much participation as is currently possible.
Now, the current level of participation is not a level playing field, and that’s a bad thing. I’m not saying “it’s all good”. I’m just saying that there is no reason that switching from providing money to publishers through subscriptions, to providing it through APCs, needs to make things worse than they currently are for the global South.
And at this point, I’ll agree with the premise at the end of your question: the goals of open access are important enough (especially for the global South!) that we shouldn’t expect solving it to remedy all inequities.
The current system is *awful* for the global South: despite some deep discounts and some foreign aid, most colleges and universities and other research institutions in poor countries cannot afford to subscribe to – and thus read – most science. Heck, most smaller universities and colleges in the *US* can’t afford to subscribe to large fractions of published science.
The scientists – and citizens – of the global South – are being denied access to vast amounts of valuable scientific discovery. Opening access will be a huge benefit – and I think that done thoughtfully, the negative impact on authors can be minimized, perhaps completely avoided.
Q: You say that a global flip will allow more scholars in the global South to read scientific results, and so they would benefit from the shift, as readers at least. This is what OA advocates have always argued, but you perhaps saw Elsevier’s recent proposal for what it calls “region-specific OA”.
This is a response to European calls for flipping subscription journals to OA and, as I understand it, would limit immediate open access to researchers based in the countries that funded the publishing charges. If that is right, then those in the South would see little or no benefit, would they?
J M-M: I don’t know in any detail what Elsevier is floating (I’d hardly call it a proposal). And given that Elsevier is about the most intransigent and unhelpful of all the for-profit publishers, I’d hardly expect proposals from them to be very interesting at this point.
They are not trying to improve the sharing of scientific knowledge in the world: they are trying to maximize their shareholders’ profits. *I* certainly don’t think we should work towards anything one might call “region-specific OA” – my goal is and has always been universal, immediate access OA. And this *especially* for the global South!
Q: So you do agree with those who argue that scholarly publishers make excessive profits from scholarly publishing?
J M-M: Yes, at least many of them, especially the large, so-called “prestige” publishers. I’ve been quite open about this in public statements.
Do we understand the implications?
Q: You maybe saw the Latin American and French rejections of the APC model. The Latin American declaration makes a point of citing David Shulenburger’s critique of flipping, which I think might have been a response to your blog post supporting it.
You said that your research specialism is the economics of information. As I understand it, Shulenburger works in the same area. What do we non-specialists make of the fact that two specialists have reached very different conclusions? Some might assume it demonstrates that no one really knows what the implications might be. Would you agree?
J M-M: Look, I don’t want to use this forum to get into a critique of the ideas of one other person – we presumably don’t have enough space to go into it in enough detail, and David isn’t here to defend himself. I *did* publish a critique of his white paper, where I explained how he got the economics wrong. David and I also appeared in a point-counterpoint session at the OAI-10 conference last summer in Geneva.
What do non-specialists make of the fact that two scholars disagree? Always a tough one. If the issue is important enough more scholars will weigh in over time and some consensus will emerge.
But the fact that two people disagree doesn’t mean that “no one really knows what the implications might be” – it could be that one of them is simply wrong. I have stated publicly – and provide the argument to support my claim – that David is simply wrong.
Q: You say that there is a significant amount of research that is of primary relevance to particular regions, and so regional solutions will meet many/most of the needs for those in the global South. Does the same argument apply to the US and Europe?
J M-M: Yes. There are journals in the US and Europe that tend to focus more on problems of interest to those areas, and are not widely subscribed elsewhere. For *publishing* (that is, from the author’s perspective), how useful these solutions are depends on how much the research authors are doing is primarily regionally relevant, and that no doubt varies from region to region.
For those publishing things that are generally useful (not regionally specific), they will want to publish in journals that are subscribed more broadly; for those doing more regionally relevant work, they will choose journals focused on the relevant audiences for that work. Both types of research happen around the world, as far as I’ve seen.
No easy solution
Q: As you will know, researchers in the South are increasingly incentivised to publish in international journals. These are usually owned by publishers in the North, and the cost of publishing in them is prohibitively high, even with pure gold OA (as opposed to hybrid OA). As you point out, some publishers offer APC waivers, but not all do, and these could be withdrawn at any time.
There are other objections to waivers too (see here for instance). If those in the South are now also incentivised to publish in international journals (which charge $1000s to publish a paper) how can pay-to-publish gold OA be a global solution? Would it not be fairer, and more sensible, to work towards a system that, to quote from The Lord of the Rings, could offer “One ring to rule them all”?
J M-M: I think I’ve mostly answered this in my answers above. There is no easy solution for global income inequality (to say the least!), and we shouldn’t try to load everything onto the way we pay for scholarly communication.
I genuinely believe the global South will be better off with a well-executed gold OA system: will have access to *much* more published research (to read), and shouldn’t have any less overall access to publishing its own research. Not *enough* access, or equal access – again, income inequality is a pernicious problem and it won’t go away. But things are much worse now with pay-to-read.
I don’t understand the last sentence in your question: I don’t know what the “one ring to rule them all” system is. But if it would produce universal open access, *and* make it easier for the global South to participate (without damaging our ability to validate and certify science), then I’d probably support it!
Q: Finally, I wanted to raise the issue of predatory publishing, which some believe those in the global South are particularly vulnerable to. I realise opinions are divided on this, but the concern is sufficiently great that last year the Indian Ministry of Human Resource Department announced [P. 15] that pay-to-publish papers will no longer be considered for faculty promotion in the 31 National Institutes of Technology.
This seems to suggest that no form of pay-to-publish model is now acceptable in India. Is this not a big impediment to a global flip? Or does India need to rethink its approach?
J M-M: I haven’t looked into the situation in India. I do think that so-called predatory journals are a problem. But attempts to exploit unwary consumers are present in *every* market solution (to the production of every good and service!). Just because someone will try to take advantage of consumers doesn’t mean we should abandon that market solution, or we wouldn’t have any markets for anything.
Right now, we have “predatory” publishers proliferating the number of subscription journals, and the number of low-quality articles published to fill them, to exploit consumers (universities and other institutional subscribers). So-called prestige publishers like Elsevier are doing this.
Because the current system separates those who produce the research (authors) from those who pay for subscriptions (libraries), for-profit publishers have found that they can vastly expand their profits by overpublishing – authors insist to their chairs and deans and provosts that we “need” to subscribe to all these journals (because they are publishing in them and want the prestige), and ever-increasing amounts of research funding is being transferred to the shareholders of for-profit publishers.
One solution – imperfect but it’s what we’ve got – is to resist subscribing to lower quality journals by using our heads and evaluating their quality.
In a gold OA world, various publishers may publish poor science because some authors will be willing to pay them for the privilege because they think it will advance their careers. We’ll have to be alert to that. We should judge the quality of journals on what they *publish* however, not on what their business model is.
We’re not *stupid* consumers – we should use our heads, and evidence, and make good decisions. And bad scientists who want to pay to publish bad science in bad journals will not succeed very well....our evaluation systems are not so broken that we can be fooled so much of the time.
So sure, some bad articles will get published, but that’s true today too.