Friday, October 27, 2006

Open Access and the Cost of Publishing

As posted to the American Scientist Open Access Forum, October 27, in response to a question about the approximate cost of the production, including
dissemination both in print and electronically.

Abstract: this post looks at the cost of publishing and dissemination in relationship with open access. The wide range of costs per article - from nothing, to thousands of dollars - is explained. A journal that relies on free, open source software, volunteer labor, and in-kind server support, may have no hard dollar costs. A commercial journal with paid editorial staff, profits and taxes to deal with, may have substantial expenditures.

The question of whether, and how much, to support open access processing fees is explored. The author concludes that researchers and universities should budget low for open access via author choice options, to allow market forces work to ensure that the transition from print to electronic, closed to open access looks for efficiencies such as automating publishing tasks, and that obsolete, print-based publishing tasks which may actually be counterproductive in the electronic environment, are dropped. Support open access publishers with true open access and reasonable costs, such as BioMedCentral, Hindawi, Public Library of Science, and many others. For journals with high processing fees and/or dubious open access status, author self-archiving is recommended as the only strategy for open access.

Full post:

This figure is highly variable, and in the process of change. With
ongoing developments in automation (publishing and word processing
software improvements, continuing decreases in per-byte costs for
storage of information), the cost of publishing an article is in a
process of decrease, and I do not believe we have yet seen equilibrium.

When considering publishers' "open choice" type arrangements -
payment for open access on publication - it is important to keep in
mind that open access does not apply to print. As long as journals
continue to be produced in print, this portion of the revenue stream
should come from print subscriptions.

The dollar costs of coordinating and disseminating a peer-reviewed
research article ranges from nothing to thousands of dollars.

The model that equates with zero costs is one based on voluntary
labor, free open source publishing software such as Open Journal
Systems [disclosure - I am on the planning
committee for the First International PKP Conference), and in-kind
support. For example, your university or university library might
provide free server space for a journal local faculty participate in.

Models with dollar costs obviously involve payments of a variety of
types. Some journals have volunteer editors, others paid. One can
purchase publishing software and server space. Some publishers
(commercial, and also some not-for-profits) look to profit from
publishing. Commercial entities also need to factor in taxes.

Overall, the costs of publishing are decreasing dramatically, due to
the technology. The costs of disseminating an electronic-only
journal post-production are close to nothing per article, unlike
print which incurs per-issue printing and distribution costs. Many
of the tasks involved in publishing are either automated, or made
much easier by automation. Think of word processing software and the
difference this means for editing; copyediting is still needed, but
no doubt overall more articles are submitted with better spelling,
punctuation, and formatting than in the past. Word processing and
publishing software can automate much of the work that used to be
done manually. For example, thanks to citation software packages,
the work of switching from one bibliographic format to another to
prepare for publication in a particular journal can mostly be done

I'm not sure how helpful this will be to your budgeting process,
Donat. Here is another suggestion: when budget for open access
dissemination, my advice is: budget low on a per-article basis. In
my opinion, this is important, partially to ensure best use of
current available funding, but more importantly, to ensure that
market factors work towards this efficiency in publishing. That is,
paying high per-article processing fees reflecting publishing
practices which should be in the process of becoming obsolete, is

One way open access can be fully met with this low budget approach:
Most journals allow authors to self-archive, and there is no cost
involved to the researchers or their institutions.

I would encourage researchers to support open access publishing as
well. There are open access publishers who provide true open access,
and do not need to charge processing fees, or who charge reasonable
processing fees (such as BioMedCentral and HIndawi; also PLoS - more
expensive, but still reasonable considering they are aiming at the
top-quality market).

Not every open choice option is true open access, and not all fees
are reasonable. If a researcher (perhaps with help from their
university library) is able to distinguish the options that really do
move towards open access, these are worth supporting. If means are
not available to deal with all the complexities, my advice is to
stick with self-archiving, using tools such as the Sherpa Romeo list
and the new addenda for authors to keep their rights from JISC,
SPARC, and others, to ensure that authors keep their rights to self-


Heather Morrison

This post reflects my personal opinion only and does not represent the opinions or policy of the BC Electronic Library Network or the Simon Fraser University Library.

1 comment:

  1. Heather - the use of the word "reasonable" is interesting in the context of your post. What seems reasonable is related to what people are accustomed to. Somehow, we have gotten used to a system where editors are paid but anonymous reviewers are not, even though both have to be highly qualified to do their job and it takes time to do it well. All of this will change as people get more options for publishing and finding the information they need to do research. You are providing a valuable service to the scientific community by reporting on the changing options.


Thank you for your comment. Comments on IJPE are moderated.