In the second week of January, the Academic Publishing in Europe (APE) meeting offered one of the first international scholarly publishing events of 2021. A panel discussion, organized by Elsevier in this all-virtual conference, brought in the views of researchers, editors, and publishers from Bangladesh, Cameroon, and Gambia on Open Access (OA).

Group of open doors with blue sky and sun

Last June, the STM Association and the International Center for the Study of Research (ICSR) produced a white paper on how to make an OA transition equitable for the researchers of Low- and Middle-Income Countries (LMICs). The study talks about the Global South’s presence in the academic and OA publishing arenas over the last decade or so, reasons behind the seen trends, and possible actions to improve South’s inclusion in the OA system. The APE panel discussion started with a presentation on this white paper to set the scene. While going through this paper, I recognized four issues, which I thought worth sharing in my remarks as an APE panellist from Bangladesh.

My first point reiterates “every country has a story” — as the white paper noted. During 2009−2018, for example, among all Research4Life countries, Ukraine was one of the highest research publication producers and Iraq showed the highest growth, while Venezuela was the only country showing negative growth in research outputs.

Earlier this month, based on the information submitted to the University Grants Commission, the coordinating body of Bangladeshi universities, a newspaper report disclosed that the country’s universities only spent 1% of their 2019’s expenses on research. Despite the poor investment, many of the 31,600 teachers of 150-odd universities of Bangladesh do conduct research and publish papers in journals published from within Bangladesh and abroad. The recruitment and promotion policies of many universities, as well as many senior academics and policy-makers of Bangladesh however, do not recognize the dynamic academic publishing world. For example, they do not distinguish between indexed and non-indexed journals, do not follow the shift towards the OA model, nor do they take any action to protect their academics from predatory journals. As a result, there is no real incentive for Bangladeshi academics to spend several months’ salary on an article processing charge (APC) to publish a paper in a Gold OA journal (which asks authors to pay an APC making an article immediately open access upon publishing) or even check if they can get a waiver being citizens of a Research4Life country.

Bangladeshi researchers, therefore, often publish their research in peer-reviewed journals published in Bangladesh. There are apparently more than several hundred journals published by Bangladeshi societies and institutions: 143 journals are hosted by the BanglaJOL; the Scopus indexes about 20; Journal Citation Reports (JCR) includes only four. Almost all journals are Gold OA with no APC (also known as subsidized, Platinum OA or Diamond OA), supported by government grants or institutional subsidy. I call this rather closed publishing practice ‘scholarly isolation’.

So, when we discuss how to include the Global South more effectively in OA publishing, we need to ask ourselves — what are the other barriers, besides APCs, stopping the South? How can we overcome those barriers? And does the South really want to break these barriers?

Some recent developments in the university ranking system, however, might be influencing the academic institutions in the South to think differently. That brings us to my second point. Bangladeshi universities are nationally criticized for not doing well on the World University Rankings of the Times Higher Education (THE). But, they in fact cannot do much about it as 60% of the ranking score is based on research and citation. In 2019, THE introduced the Impact Rankings, which measure universities’ contributions to the 17 Sustainable Development Goals (SDGs) to rank them. This new system scores research, stewardship, outreach, and teaching almost equally. While it emphasizes research much less than THE World University Rankings, the new system only considers articles published in Scopus-indexed journals. That has made many Bangladeshi universities encourage their academics to publish in Scopus journals, which disqualifies almost all journals published in Bangladesh.

Publishing research in a recognized journal is not enough to get points for the Impact Rankings; it has to contribute to specific SDG(s). This demands a major shift in the ways we design our research, get our funding, and write our articles. The STM Association-ICSR’s white paper recommends supporting the publication of research relevant to the SDGs, which may create new OA opportunities for Southern researchers. But, a shift in mindset might be a bit more difficult than shifting to Scopus journals.

My third point is on the OA model itself. The white paper assumes the Gold OA model as “the” model, and then explores avenues to bring the Global South in. In 2018, the LMICs constituted 5.5% of world’s publications indexed in Scopus. In the same year, 23% of articles published by these countries were in OA journals, while 75% were in subscription journals. While looking into how the South is doing with OA, we also need to convince OA journal publishers to engage with the South more — not only working with them as authors, but also as peer-reviewers and editors. We also need to look into the wide range of APCs available: from zero APC in subsidized Gold OA journals to US$ 9,900 to publish in Elsevier’s Cell (Editor’s Note: Please see comment from Elsevier’s Anne Kitson below). Transparency in determining APCs might clarify how much of it is actual expense, how much is profit, and how much the author is paying for the brand value.

My final point is that when we talk about OA, we essentially talk about journals published by big, commercial publishers from the Global North. We need to bring the OA journals of the South to the discussion. It could be argued that the editorial process and quality of these journals are not up to the mark, these journals are not indexed, they do not have “impact” as defined by Impact Factor or CiteScore. But, they do publish data and information important for a particular country or a region, and these data could be crucial in crisis moments, like pandemics, natural calamities, and climate change.

The white paper recommends investing in the capacity of the researchers, reviewers, and editors of the South. Organizations, like INASP have been doing that over the last couple of decades. It also designed the Journal Publishing Practices and Standards (JPPS) and worked with several hundred African, Asian, and Latin American journals to introduce a three-star rating system. Now the question is how big journal publishers, for-profit and not-for-profit publishers, publishers’ associations (e.g. ALPSP, STM, and SSP), and global networks working on research communication (e.g. Research4Life), can help small national and regional Southern journals to improve their standards. Without that, OA discussions and actions to create a level playing field for the South will remain incomplete.

To make effective collaborations in academic publishing between the North and the South, we need to identify champions from the Global South who are not only excellent researchers, but who also understand and think about research communication in light of the whole research system and development. They are aware of the major changes ongoing in academic publishing and can act as conduits for the wider academic community. They are the ones who will work with the universities and government entities and guide improved policies and practices, which will make Southern research and research communication impactful, not just some text in academics’ CVs.

Haseeb Irfanullah

Haseeb Irfanullah

Haseeb Irfanullah is a biologist-turned-development facilitator, and often introduces himself as a research enthusiast. Over the last two decades, Haseeb has worked for different international development organizations, academic institutions, donors, and the Government of Bangladesh in different capacities. Currently, he is an independent consultant on environment, climate change, and research systems. He is also involved with University of Liberal Arts Bangladesh as a visiting research fellow of its Center for Sustainable Development.

Discussion

23 Thoughts on "Open Access and Global South: It is More Than a Matter of Inclusion"

Australia, New Zealand and South Africa are in the Southern Hemisphere. Afganistan is in the Northern Hemisphere. Is it time we came up with a better and more diversity-aware way to describe lower-income countries than the clichéd term ‘Global South’?

Factfulness by Hans Rosling offers some better alternatives, for example.

Thanks for this thoughtful and illuminating post, Haseeb. You’ve really underlined the complexity and challenge involved in our vision of creating a system that is open and equitable for all. Too often, the global North thinks of LMICs as simply a source of papers and/or subscriptions/APCs. Instead, we need to be much more focused on understand the dynamics country by country, and what the local barriers and needs are. This is embedded in our approach as we expand globally at PLOS but clearly requires a much bigger systemic approach if we’re to make real progress towards that vision.

Thanks Alison, for sharing your thoughts on enhancing inclusivity in OA policies as well as practices.

Useful article – thank you. I fully agree with the need to build capacity among national and regional journals that publish research useful for the populations of those regions. The African Journals Partnership Program has been running for many years, whereby high-impact medical journals in the USA and UK pair with less established journals on the African continent to provide advice and support in developing journal policy and practice. Indeed some of those African journals (eg, the Ghana Medical Journal) are now indexed in Scopus and doing very well. More of this around the world should be on every publisher’s agenda.

Thank you for sharing recent cooperation between African journals and American and British journals. These are really promising and should be practiced by others. Thank you.

Interesting reading, thank you.
If we want to achieve the SDG in our developing regions we need to value contributions published in quality journals indexed by Scopus and WoS and also contributions in quality journals indexed within our regions. Examples in my region, Latin America, are journals indexed by SciELO and Redalyc.

Thanks Haseeb for this excellent post, one additional resource not mentioned here who are doing great work in these regions is the DOAJ https://doaj.org/ supporting the journal indexing process, helping educate publishers/journals and editors around standards, and with a very effective group global ambassador within countries – I was moderating a Charleston webinar session that I thought was excellent to help define excellent how DOAJ really is a diverse and inclusive collection of OA journals from around the globe, here’s the recording if interested, and happy to chat further offline too, this is an important topic for sure https://www.charleston-hub.com/extending-the-reach/

Thanks Adrian, for mentioning DOAJ and sharing the link to Charleston webinar recording.

Thanks for the insightful article Haseeb, there was one point that I thought was important to clarify. You mentioned the broad range of APC costs that researchers may have to pay to publish their work (from zero on Gold OA up to $9,900 on Cell). I realise this example was included for context, but my concern is that this could create some confusion as it does not mention Elsevier’s continued support for the Research4Life programme.

In reality, Elsevier waives 100% of the open access APCs for all 69 Group A countries (including Bangladesh) in the Research4Life program and reduces the APC by 50 per cent for the remaining 57 Group B countries.

So using that figure, even as an outlining example to highlight the differences in costs, could continue to confuse and dissuade Global South authors from publishing their research in the most high profile journals. It would be a real shame if authors from the Global South did not attempt to get their research published in journals such as Cell if they incorrectly assumed they could not afford to.

Thanks very much, excellent discussion. We need to be fully aware of the range of journals on offer. And treat them ethically, using social justice criteria. Non-Western outlets would then get better billing.

In my social science listing, https://simonbatterbury.wordpress.com/2015/10/25/list-of-decent-open-access-journals/ I cite many english language journals, reputable ones only, platinum OA journals with APCs of $0 [mostly] up to $500. Many high-quality journals are listed from non-western countries. It may be of some help.

I think Latin America with its multilingual OA, free academic-led journals and journal aggregators like SCIELO https://scielo.org/en/ is not only catching up with the West and providing publishing alternatives, but doing so on its own terms and in its own way. The DOAJ provides a good service mainly in English to link to journals but it not a publisher.

Also if universities sign up to DORA, largely a Western initiative at present, then they have to retool their internal procedures to focus on assessing academic worth based on quality, not place of publication. That is something for Bangladesh universities to consider. It does not address the Scopus listing issue though. DORA helps recognize your “dynamic academic publishing world”. Its core argument is that signatories should “eliminate the use of journal-based metrics, such as Journal Impact Factors, in funding, appointment, and promotion considerations;” and “assess research on its own merits rather than on the basis of the journal in which the research is published”. This would require better university assessment procedures https://sfdora.org/read/.
• On the true cost of publishing OA articles, this new paper by Grossman and Brembs https://f1000research.com/articles/10-20/v1 really begins to answer that – about US$400 on average with a $200-$1000 range. Cell should be charging something in that range. Anything above the average cost for different journal types is profit, as the almost criminal APC charges applied by the big 5 publishers illustrate.

“Competing interests: Björn Brembs is an academic editor with PeerJ and used to be academic editor for Frontiers and PLoS. Alexander Grossmann has been Publishing Director at Wiley, Managing Director at Springer, Vice President Publishing at De Gruyter, and is currently Professor of Publishing Management at HTWK Leipzig and Co-Founder and President of ScienceOpen.” I think they have data.

Academic editors are usually (or at least are in an ethically run publishing houses) separated by a firewall from financial matters of the journal.

I think they have data.

Would have been nice if it was included in the manuscript. The word “estimate” or “estimation” appears in the paper 18 times. “Assumption” or “assume” 13 times.

“Would have been nice if it was included in the manuscript.”

lol

we hid it extremely well – under the apparently completely mislabeled section “data availability” and, a second time, in the methods section.

Looks like being able to search for keywords is not the same as being able to read 🙂

It was pretty straightforward to collect that data: take the steps it takes to publish a paper, add the currently available market rates for each step up and add fixed costs according to various scenarios. It’s under review, so if you find errors in that process, we can still fix the manuscript. After more than a year online as a preprint on PeerJ (8500 views, 4300 downloads), nothing substantive so far.

P.S.: Thanks for the suggestion: we have now replaced all 9 of the 18 instances of “estimat*” with “calculations” where we were referring to our own calculations of cost. There was no reason to change any of the “assum*” instances.

Honestly, I haven’t read the paper in detail nor given it much thought. On a cursory reading, the numbers offered differ considerably from what I’ve seen for the many journals I’ve worked with over the last 20 years. I also strongly disagree with your assumptions about what is considered a publishing cost and what is excluded, as well as your assumptions about what readers and authors value and want included.

But none of that really matters. A theoretical paper about what publishing should or might cost is of little value, because publishing is not a theoretical activity. What matters is what publishing does cost. So to prove you’re right, you need to do the actual experiment. Start a journal, run it under the conditions and costs you propose and prove that this is the most effective and successful way to publish research.

Well, J Pol Ecol which I have edited since 2003 comes out on the low side of their estimates, largely editorial labour cost, which is pretty easy to calculate from known salary costs [while we have any salary as academics… thousands of uni social scientists are being made redundant as we speak across the world because of covid/lost fee revenue, especially in Australia where I work].This would be lower if we were based in the Global South, or under/unemployed. Other journal costs are the same as in the F1000 article but do not amount to much. However we have variance, because some articles need more time. We have just reduced costs by adopting APA referencing, because our previous style was unavailable on commercial software and we did a lot of that by hand.
Interestingly, the journal Conservation and Society, WoS and Scopus listed and OA, has regretfully gone from no APC to a US$600 charge per article for those who can afford it. It is published wholly in India. This is to pay editorial costs that were previously covered by an institution [I know the debate and some of the people]. The decision is explained here. http://doi.org/10.4103/cs.cs_240_20 .

That’s two journals, but laid bare, it still looks to me for-profit large publishers are overcharging substantially at the expense of authors and readers. I think the weight of evidence is growing. I have read tens of articles on this site justifying their actions so I know it is not that simple. Yes they employ thousands of people which is nice, handle data, etc. but annual reports still reveal high exec salaries, and fantastic profit margins. I guess us academics have been making this point for decades, [the Manifesto https://doi.org/10.21428/6ffd8432.a7503356%5D and it is finally beginning to get through – publishing can be socially just, and cheap. We can do a lot of it ourselves, or with cognate institutions, even if a bit more support would be welcome.

lol, just check Alexander’s bio for how many journals he has run (and is running) at that cost.

This comment demonstrates why I barely, if ever come over here any more:
a) Either you really don’t know the publishers who are publishing many journals at exactly these costs despite them being very prominent (and listed in our paper, btw), then you’re not competent enough to be taken seriously, or
b) you do know these journals and then your comment is just completely insincere trolling

So why should I ever come back and waste my time here? Bye bye!

P.S.: Just today, in addition to what Simon Batterbury wrote, a random pick from many, many possible picks over the last decade, $271 per article: https://twitter.com/scholasticahq/status/1361937181498105856
So the experiment has already been performed many times over and validated everything we calculated. Simon is right “for-profit large publishers are overcharging substantially at the expense of authors and readers. I think the weight of evidence is growing.”

There are indeed a lot of low-cost options already on the market for researchers, and yet it doesn’t seem to have slowed submissions to Nature or Cell. And I think this points to the misconception that pricing is based on costs, rather than value. Authors and readers perceive a value in a certain journal, and if the price (for an APC or a subscription) doesn’t align with that value, then the journal won’t publish many papers or have many readers. If journals are indeed “overcharging” as you note, it should become evident in the market, but every report I read from the big commercial publishers is that their marketshare and profits continue to increase every year. Over time, if you are correct, then we will see this change, but current trends suggest there is value perceived by authors and readers that is not accounted for in your analysis.

From the academic side, the ‘value’ of a Journal is actually set by hiring and promotion processes in unis as well as by individual academic’s valuation of their own self worth/success which is tied to those processes. The way these dominate decisionmaking cannot be regarded as particularly desirable. The DORA decoration is one effort to take a different view of prestige and value and a lot of universities are now signing up, including mine. As long as publishing in journals like Cell or Science is seen as an amazing achievement, regardless of the quality of the article, we will see little change. Having established just above that many commercial APCS are much higher than net article publication costs, the next question is whether the exploitation of scholarly value pursued by the large commercial publishers is ethical. My view is that it is not when it generates such enormous profits, as shown by the way these are lauded in annual reports as creating shareholder value and exec remuneration. The high APCs accompany the capitalist publishing firms, as many authoers have shown eg https://doi.org/10.1080/00330124.2016.1229624. Where you see a value driven market [the sort that economists also love] , I see rampant exploitation of poorly paid scholars particularly in the Global South [Cannot afford the APC’s, rely on research4life or discounts to meet APCs, or cannot publish a really good idea because it is not worthy of so-called western publishing norms]. Uni libraries cannot afford the remaining journal subscriptions that we would like them to have. Profit driven academic publishing is in many respects rotten to the core for these reasons. So we can either do it ourselves, form cooperatives as the radical Open Access movement does, or find better ways to pay the publishing workforce where this is deemed necessary in not for profit settings , through funder and university core fund transformative or other agreements only to publishers operating in an ethical way.

While I largely agree with your overall conclusions, I have some problems with how you’re justifying them.

First I think your view of the value of journals is something of an oversimplification. Yes, absolutely, the prestige offered by publishing in a particular title is a key factor in journal choice. Paula Stephan explains the reputation economy well in her book (reviewed here https://scholarlykitchen.sspnet.org/2017/06/20/revisiting-book-review-economics-shapes-science-paula-stephan/ and interviewed here https://scholarlykitchen.sspnet.org/2012/04/11/interview-with-paula-stephan-economics-science-and-doing-better/). But Impact Factor is far from the only determining factor in journal choice. In study after study, researchers nearly always put factors like reaching the right audience and rigor of the review process at the top of their reasons for journal choice. Are they lying or do these things actually matter? There is certainly an argument for a mixed population of motivations – the growth of megajournals, which have fewer peer review criteria and often have higher Impact Factor rankings than niche specialty journals would suggest that for some part of the community you are correct. But that’s not the entirety of researchers, and for many, things like the reach into the right community, the marketing of the publications, the ease of submission, the responsiveness of the editorial office, and the structure of the peer review process (e.g., eLife, F1000), among other factors, play a role.

DORA is now nine years into its existence. How effective has it been in changing the career and funding structure of academia? With many universities signed on yet still continuing to explicitly list journal prestige as criteria in hiring and tenure decisions, how effective a tool is it? With the largest and most profitable commercial publishers now signatories that continue to market their journals through promoting their Impact Factor rankings, how effective a tool is it?

To get back to the economics, again, you’re taking a non-peer reviewed study as fact, and ignoring the transparent reporting from actual publishers. Do you think EMBO, PLOS, and eLife are all lying when they make public their costs for publication? Are these reports less credible than a set of estimations from a pair of researchers who exclude significant expense categories from their calculations? Why does one set of numbers count but the other doesn’t?

It’s worth noting that at one point, Hindawi, which had relatively low costs and low APCs, was running at a greater than 50% profit margin, more than the most rapacious of the big five commercial publishers (https://scholarlyoa.com/hindawis-profit-margin-is-higher-than-elseviers/). Meanwhile, EMBO’s $5200 APC has to be subsidized with subscription funds to break even. Clearly the raw number of an APC is not a measure of how exploitive it may be.

Should it be considered a problem if APCs are higher than publication costs? Generating a surplus is absolutely essential for the continuing existence of a publisher, whether for-profit or non-profit. How can one continue to experiment with and maintain new technologies or features for authors and readers or weather difficult economic periods without capital to invest and sustain?

How much profit is acceptable, and how do you measure that number? Is it a raw amount per journal or a percentage based on the costs incurred by that journal? If the former, once you hit that amount, there’s no further motivation to do anything better. If the latter, then you’re selectively driving up journal costs because higher costs would result in higher profits.

All that said, I do agree with your larger points. Subscription models shut out readers, and APC-based OA models shut out authors. Both cause serious inequities. I also strongly agree that the research community should own and control their own system of communication. See also this recent post about how libraries and universities could have more influence on the way publishers behave and align themselves with the values of their customers ( https://scholarlykitchen.sspnet.org/2021/01/13/guest-post-evaluating-publishers-as-partners-with-libraries-and-higher-education/). But to build a better system effectively, one has to rely on accurate data and understand both what the community wants from that system and what the real world costs of maintaining and improving that system are, rather than making estimates about how things “should” work.

And getting back to my earlier comments in this thread, if a bare-bones, stagnant set of journals is highly attractive to authors based solely on low prices, then these should dominate the market and prove extremely successful. The only way to prove that’s the case is to do the experiment.

Comments are closed.