The economic case for open access in academic publishing

As hyperbolic as it may sound, academic publishing is the curator and guardian of the accumulated scientific knowledge of the human race, 1600 to present. It is also a cornerstone of modern science, preferentially selecting well-executed research through the peer review process. However, academic libraries are facing decreasing budgets, and even highly ranked universities are having to cut back on journal subscriptions. Since these subscriptions account for up to 75 percent of publishers’ revenues, the entire system is feeling the pressure of the economic crisis.
But that’s not the only source of strain. As content migrates to the Web and becomes increasingly difficult to control, the “walled garden” subscription business model used by the publishing industry is facing the same issues as other major content providers. This article examines the academic publishing industry and looks at the impacts of business models from both an academic and economic standpoint.
Academic publishing is big business. Yearly revenue estimates for the industry top $12 billion worldwide. If you account for the time that researchers spend preparing and reading journal articles, the total cost tops $100 billion per year. Academic publishing is also spectacularly profitable; 2009 earnings reports from two of the largest publishing houses, Elsevier and Wiley, show that the profit margin for scientific journals is 30-45 percent and profits from scholarly journal business units have increased every year since 2005.
When many people see these numbers, they naturally ask whether it would be better to adopt online-only publishing and cut commercial publishers out of the process. After all, it’s the scientists who perform the research, write the articles, edit the journals, and perform peer review. Article authors also usually pay “page charges” of $600-$2,000 to either the publisher or the academic society that owns the journal. Ultimately, it’s these same scientists who pay the subscription fees to academic publishing houses to access others’ work.
Surely there is no need for a publisher when researchers do the vast majority of the work and pay for the results. Further, articles can be archived online, eliminating physical “dead tree” editions and the costs of printing them. The real costs of publishing
Even for online-only content, however, the real costs of publishing a scientific article are substantial. First copy costs (i.e. the amount of money spent on an article before it is published) include services such as content management systems, article submission and review, bandwidth costs, typesetting services, editorial services, editor pay, and business costs (marketing, etc.). In other words, the publishing houses do a lot more than just print paper copies of the articles.
While accurate costs from the publishing houses are nearly impossible to find (they consider their costs proprietary information), several recent studies put the first copy costs, before considering printing costs, at $2,000-$15,000 per article. Many variables impact first copy costs, but the largest is selectivity. If a journal is highly selective, it must pay for peer review of many articles for each article it accepts. For example, the estimated first copy costs for Nature and Science are $10,000 to $15,000, while the first copy costs for highly field-specific, low-selectivity journals is closer to $2,000.
With these numbers in mind, it should come as no surprise that both academics and publishing industry reports predict that moving to online-only content will reduce first costs by only 10 to 20 percent. Thus, simply moving all content to the Web will not provide vast savings to the academic publishing industry, so discussions about the future of academic publishing have shifted to content management and new economic models for the industry.Open access versus the walled garden
Due to recent US legislation mandating that articles resulting from certain types of government support be freely accessible, open access publishing has been the focus of most current debates.
Publishers receive 68 to 75 percent of their revenue from academic library subscriptions. Corporate subscriptions account for 15 to 17 percent of revenue. This revenue goes largely to the first copy costs, and these costs are the same for both traditional and open access content. Thus, the revenue stream is critical for hard copy, online only, and open access content. Any system that eliminates the need for subscriptions, like open access, would therefore force academic publishers to completely change their business models. For this reason, the publishing industry is extremely worried about the rise of open access publishing.
To offset revenue losses from open access articles, journals generally charge 2 to 10 times more if authors want or are required to use open access. For authors who are not legally required to use the open access option, adoption is negligibly small—a fact that the publishing industry uses to prove that open access is not important to researchers.
From my own experience in research labs, this line of reasoning is profoundly ignorant of the cost pressures and budget constraints of modern labs. Researchers always want their work available to the widest possible audience, but the choice between $600 in author charges to place an article behind a paywall and $5,000 for open access is a no-brainer, especially when authors know that the vast majority of researchers have adequate access through the current subscription systems. The state of open access today
As an alternative to the current system, open access journals like PLoS ONE have begun to crop up in a variety of scientific disciplines, some with substantial notoriety. Several studies have found that open access articles have higher citation rates than similar articles in subscription-based journals. It is, however, unclear whether the higher citation rates are due to open access or other factors like rapid online publication (a feature many subscription-based journals have emulated).
Some open access journals reduce costs by being less selective than subscription-based journals and eliminating the printing process. They generate revenue by selling ad space and charging authors. While these journals have experienced moderate success in the short term, many have been forced to raise author page charges over time, making it harder to attract high quality articles and jeopardizing the viability of the journals. Economic impacts of moving to open access worldwide
A 2008 report from the Joint Information Systems Committee (JISC) modelled the economic impact of the current publishing system, an all open access system, and a self-archiving system like Archiv.org. The most significant impact of shifting to open access systems is shifting the publishing cost burden from subscribers to article authors. Under these scenarios, highly productive research institutions see costs increase while teaching institutions have substantial savings.
The same observations hold true on a national scale. The cost of scholarly publication increases for countries that produce large numbers of articles per capita.
Using a complex economic model, the JISC report finds that the benefit to cost ratio for switching to open access publishing in the UK is 13.9—in other words, the financial benefits significantly outweigh the costs. Moving the entire industry to a self-archiving system, like Archiv.org, would produce a benefit to cost ratio of 122. However, self archival systems like Archiv.org have no peer review, so switching to a similar system may have negative consequences for article quality which were not accounted for in the model.
It is important to note that these findings were likely worst-case scenarios in a high article output country. A similar study in 2009 by the Research Information Network found that switching to open access journal publishing would produce nearly $1.5 billion in savings worldwide.
The publishing industry has criticized the economic model used in the JISC report, stating that it underestimates the costs of overhauling the academic publishing system, underestimates the efficiencies of the current system, and that many of the cost savings would lead to job cuts for librarians and publishers. Many of these criticisms, however, appear to be well off the mark and show a profound misunderstanding of the models in the JISC report. Further, the economic model used in the JISC report is available online and the publishing industry is free to adjust the assumptions and show what they consider to be a more realistic simulation. It seems rather telling that no such analysis has been reported.
Ultimately, I believe the academic publishing world will, and should, slowly shift toward open access, but the transition will be ugly. The issue boils down to a classic problem in economics: the tragedy of the commons. While the publishing industry and researchers continue to act in their own short-term self-interest by continuing the status quo, we are slowly heading toward an untenable situation where the people producing research papers will not be able to afford to access them.
At the same time, authors and researchers need to do a better job of understanding their rights under the current system. In almost all cases, commercial journals allow article authors to host and distribute pre-print copies (versions of the article either before peer review or without the layout used by the journal). Authors are therefore free to archive and distribute versions of their articles, and legal access for researchers without journal subscriptions is usually only an email away.