About Open Access. Or What Is Really Open in Open Access Journals

Today many researchers publish their papers in Open Access (OA) journals. Sometimes they do this because publishing there is faster and more convenient than in subscription-based journals. Sometimes because they want their paper to reach as many readers as possible and, as a result, get more citations.

This raises a question: how reasonable is this strategy? Are papers published in OA journals really cited more often? And is it really easier and faster to publish there? It turns out that these questions interest not only me. There are already many studies on this topic. This post is based on several of them.

What Is Open Access and What Types Exist

First, it is important to distinguish between different OA models.

There are so-called diamond OA journals, which make articles freely available to everyone and do not charge authors any publication fees. Usually, these journals are published by large research organizations with stable funding and grants.

There are also gold OA journals. They also provide immediate open access to articles, but they charge authors an APC (article processing charge) for publication.

Finally, there are hybrid OA journals. They try to combine two models. On the one hand, they continue to work by subscription. On the other hand, authors can pay an APC, and then their article becomes open access. Not surprisingly, this model raises the most questions.

How Open Access Journals Appeared

In general, it is argued that the idea of OA journals emerged during the so-called serials crisis. This crisis was caused by the rapid increase in subscription prices for academic journals, which grew faster than the budgets of universities and research institutes.

With the spread of the Internet in the early 1990s, it seemed that there was a real opportunity to distribute scientific knowledge freely and provide access to both new and old articles. The first OA journals appeared at that time. They mostly relied on unpaid work of editors and volunteers and used university resources and grant funding.

This situation changed in the early 2000s, when large publishing houses entered the OA market. BioMed Central, founded in 2000, is often mentioned as the first professional publisher focused on OA journals. It launched many journals in biology and medicine and introduced APCs in 2002. Initially, the standard fee was around 500 US dollars, but by July 2005 it increased to 750 British pounds. After the success of BioMed Central, other publishers followed. For example, Springer began to introduce OA options in 2008 (Borrego, 2022).

Paying to Be Published: How Strange Is That?

When APCs were first introduced, editors explained this model by saying that authors benefit from publications, so it is fair that they pay journals in some form. It was also argued that this model is cheaper overall than journal subscriptions and that authors gain more control over their articles. In the traditional model, authors often cannot freely distribute their work for a certain period, usually around two years. In addition, supporters of APCs noted that authors had already been paying journals for things like color figures or extra pages.

Interestingly, many authors initially refused to publish in OA journals because they were not willing to pay for publication. For example, in 2005 only about half of surveyed authors said they were willing to pay APCs (Nicholas & Rowlands, 2005). Despite this, the number of OA journals charging APCs continued to grow. In 2009, there were about 2,000 such journals, and by 2016 their number had increased to almost 10,000. At the same time, the number of articles published in these journals grew from 8,000 to 45,000 (Björk, 2017).

According to data from 2013 (yes, more than 10 years ago, but still informative), 28% of more than 9,000 journals listed in the Directory of Open Access Journals (DOAJ) charged authors publication fees (Kozak & Hartley, 2013). This share was much higher in medicine (47%) and natural sciences (43%) than in the humanities (4%).

More recent data show that only about 30% of OA journals charge APCs, but these 30% publish around 56% of all OA articles.

It was expected that the introduction of the hybrid model would reduce subscription prices. However, available data do not support this. According to Besancenot and Vranceanu (2017), the shift to hybrid OA increased the revenues of large publishers and raised costs for authors. On average, hybrid journals charge even higher APCs than gold OA journals.

How Much Does It Cost to Publish?

APC amounts vary greatly between journals. Many studies show that APCs are higher in high-impact journals published by large international publishers. Importantly, publication costs have been rising steadily and much faster than inflation.

In 2005, the average cost per article paid by European institutions was 858 euros. By 2018, it had increased to 1,600 euros. If adjusted for inflation, the cost would have risen only to about 1,100 euros (Khoo, 2019). According to the same study, between 2012 and 2018 APCs increased by 25% to 60% across different journals.

Data from the University of Cambridge show that between 2015 and 2020 more than 30% of grant funds allocated for OA publications were paid to Elsevier. The average cost per article was 3,302 pounds, and the maximum cost reached 7,320 pounds (which was over 9,300 US dollars at the 2020 exchange rate).

The hybrid OA model continues to expand. For example, in 2020 Nature announced a move to a hybrid OA model, with 33 journals charging APCs. At the same time, subscription prices for many Nature journals remain high.

Criticism of the OA Model

The OA model with APCs is often criticized for creating inequality between early-career and senior researchers and for putting researchers from poorer countries and institutions at a disadvantage.

However, a study by Kieńć (2017) showed that authors from countries with GDP per capita below 18,000 (defined as “peripheral countries”) publish in OA journals as often as authors from wealthier countries. The important detail is that they mostly publish in journals within their own countries. These journals usually have lower impact factors and lower APCs. Another study published the same year showed that authors from developing countries are disproportionately represented in lower-ranked OA journals.

Peer Review Quality

The quality of peer review in APC-based journals has also been questioned. Several experimental studies addressed this issue. One well-known example is the study by Dell’Anno et al. (2020).

The authors submitted a “bait manuscript” to 73 economics journals to test whether there were differences in peer review between journals that charge APCs and those that do not. The manuscript had obvious flaws: methodological errors, problems with the literature review, and completely fabricated data.

As the authors wrote, the manuscript contained at least 71 errors that a competent reviewer should have identified and recommended rejection. These included poor English, wrong derivatives, reversed interpretations of results, and incorrect conclusions about statistical significance.

The results were striking. Among journals that did not charge APCs (39 journals), 95% rejected the paper in the first round, and the remaining 5% requested revisions but rejected it later. None accepted the paper. Among journals that charged APCs (34 journals), 23.5% accepted it in the first round, another 23.5% requested revisions and then accepted it. Overall, more than half of APC journals accepted the paper.

In a second round with 30 additional journals, the results were similar. None of the non-APC journals accepted the paper, while 57% of APC journals did. Based on this, the authors concluded that peer review quality in APC journals is problematic.

Open Access and Citations

One common argument in favor of OA is that OA articles are cited more often. However, the link between OA and higher citation rates is not well established.

A review of 134 studies on OA and citations showed mixed results (Langham-Putrow, Bakker, & Riegelman, 2021). About 47.8% of studies found a citation advantage for OA articles, 27.6% found no advantage, and 23.9% showed benefits only in specific fields. Overall, it seems that not all OA journals provide the same citation benefits, and not for all disciplines.

Attitudes Toward Open Access

Finally, what do authors themselves think about OA and APCs? One study examined researchers’ attitudes toward OA journals and publication fees. It found that many authors see APCs as a global threat to science rather than as a personal career problem. Early-career researchers and scholars from low- and lower-middle-income countries expressed the most negative views. Differences were also found across disciplines: researchers in the humanities and social sciences were more critical of APCs than those in STEM and life sciences.

Overall, among researchers who study the economic and structural effects of OA publishing, negative views of APC-based and hybrid models seem to dominate. One article summarizes this view well:

“In short, we are getting fleeced. The major scientific publishers enjoy profit margins in excess of 30 percent. Such profits are far above the average for Fortune 500 companies. Publishers get rich on the backs of underfunded academic libraries and the unpaid labor of academics who act as editors, reviewers, and authors. This system is unsustainable” (Suarez & McGlynn, 2017).

Despite all these problems, many authors have little choice and end up paying APCs. Given the pressure to maintain high publication output and long review times (especially in the social sciences), OA with APCs sometimes remains the only realistic option.

A personal note

There is one more thing that really annoys me. The American Psychological Association (APA) still restricts access to many old articles. I can understand why very recent papers might be behind a paywall, for example articles published one or two years ago. But I honestly do not understand why access is restricted to papers published decades ago, for example, in 1974.

Science does not develop only through new publications; it also relies on older work, theories, and debates that form the foundation of current research. What is the point of closing access to such old studies? How can we talk about the development of science if publications on a topic remain available only to a small group of people who have institutional access? This makes scientific knowledge unnecessarily exclusive.

Reference:

Borrego, Á. (2023). Article processing charges for open access journal publishing: A review. Learned Publishing, 36(3), 359-378.

Nicholas, D., & Rowlands, I. (2005). Open access publishing: The evidence from the authors. The Journal of Academic Librarianship, 31(3), 179-181.

Björk, B. C. (2017). Growth of hybrid open access, 2009–2016. PeerJ, 5, e3878.

Kozak, M., & Hartley, J. (2013). Publication fees for open access journals: Different disciplines—different methods. Journal of the American Society for Information Science and Technology, 64(12), 2591-2594.

Besancenot, D., & Vranceanu, R. (2017). A model of scholarly publishing with hybrid academic journals. Theory and Decision, 82(1), 131-150.

Khoo, S. Y. S. (2019). Article processing charge hyperinflation and price insensitivity: An open access sequel to the serials crisis. LIBER Quarterly: The Journal of the Association of European Research Libraries, 29(1), 1-18.

Kieńć, W. (2017). Authors from the periphery countries choose open access more often. Learned Publishing, 30(2), 125-131.

Dell'Anno, R., Caferra, R., & Morone, A. (2020). A “Trojan Horse” in the peer-review process of fee-charging economic journals. Journal of Informetrics, 14(3), 101052.

Langham-Putrow, A., Bakker, C., & Riegelman, A. (2021). Is the open access citation advantage real? A systematic review of the citation of open access and subscription-based articles. PloS one, 16(6), e0253129.

Segado‐Boj, F., Prieto‐Gutiérrez, J. J., & Martín‐Quevedo, J. (2022). Attitudes, willingness, and resources to cover article publishing charges: The influence of age, position, income level country, discipline and open access habits. Learned Publishing, 35(4), 489-498.

Suarez, A. V., & McGlynn, T. (2017). The Fallacy of Open Access. Washington, DC: The Chronicle Of Higher Education.