The purpose of this post is to provide some important context for understanding OA publication charges. A key point that I would like to highlight is that the OA APC, in the sense of a single number charged for each article published, is a myth. It is important to understand this because  OA journals are obviously conducting some real-world experiments that have the potential for beneficial results for high-quality, sustainable open access publishing, and providing discounts that may be needed by some authors and OA APC payers.

Pricing is often tailored to reflect the work involved in publishing an article. A well-written article that is submitted in good shape with little need for language editing or copyediting with camera-ready graphics is less work to publish – this should, and sometimes does, cost less. There are a variety of discounts reflecting contributions to the journal community; discounts for editors, reviewers, society or association members, loyalty discounts for repeat authors. Details about variations on APC pricing can be found in these posts by Guinsly Mondésir (Version en français / English version).

The “average” cost to pay-to-publish an article in a fully open access journal that we found in 2015 (ignoring $0 APC) ranges from an estimated $250 USD for a journal using OA page charges based on an estimate of 9 pages, to a fairly consistent median of $800, modes of $600 to $800, to averages or median numbers from $858 (weighting results to include smaller publishers) to the overall average of $998 to $1,370 for the same set of journals excluding $0 to a mode of $2,154 after removing the “Hindawi factor” (The practices of this largest OA publisher by number of journals skews the sample).

Even this wide range of “averages” conceals the full complexity involved with deciphering publisher pricing and with translating currencies into USD. A set of prices, gathered on the same date in 2015, would yield different averages if the currency calculation were conducted on a different date. In gathering the data we had to make many tough decisions about the “original” currency, because a number of publishers provide pricing in several different currencies. It is harder than one might think to decide on which is the “real” price, for example when differential pricing is provided based on the author’s location. Having a single APC would be simpler, but not necessarily better if it means a loss of a discount for authors or payers who could really use them or if it eliminates an incentive to streamline the process of publishing itself in the process of transition to OA.

With this important caveat we will now present our preliminary quantitative analysis of this mythical variable here.

Morrison, H. (2015). The mythical OA article processing charge. Sustaining the Knowledge Commons / Soutenir Les Savoirs Communs. Retrieved from https://sustainingknowledgecommons.org/2015/12/03/the-mythical-oa-article-processing-charge/

Update and correction December 2019: this early analysis did not take into account publishers and journals that had been removed prior to the beginning of our SKC APC longitudinal study. In 2019, we completed the sample by adding these journals to obtain a more complete picture of the status and APC changes. In brief, from 2010 – 2019 there was an apparent substantial attribution rate of 26%, however this appears to be an anomaly reflecting several new commercial APC based publishers (Bentham Open, Hindawi, Frontiers, BioMedCentral, and Dove Medical Press) that followed a start-up strategy of starting with a large number of journals to cover a broad range of disciplines, then retiring titles that were not successful. For the latest analysis of the 2010 dataset, see:

Analysis of the current status of journals sampled in 2010 (by Solomon & Björk) and/or by the SKC team in 2013 or 2014. In brief: the attrition rate for journals previously sampled is quite low, with 2% of the 1,559 titles previously sampled either apparently or actually no longer publishing. Most of the journals no longer active (34 of the 37), not surprisingly, are from the 2010 sampling. The vast majority of journals found to have APCs or APPCs (article page processing charges), over 90%, clearly had APCs in 2015. A noticeable number of journals previously identified as having publication charges either do not have publication charges, or cannot be confirmed as having publication charges today. The following chart and table provide additional detail. For more detail and documentation on how this data is calculated, see the open access article processing charges dataverse – under the Open access article processing charges longitudinal study 2015 preliminary dataset the data file is called DataSet – Main v.12 current status of previously sampled journals.csv and the accompanying documentation PDF file is called OA APC study 2015 current status journals samples 2010 – 2014 documentation .pdf

Citation: if referring to updated information on the 2010 dataset, please see and cite the 2019 post, not this one which will be retained for historical purposes:

Morrison, H. (2019). 2010 – 2019 APC update. Sustaining the Knowledge Commons. Retrieved from https://sustainingknowledgecommons.org/2019/11/26/2010-2019-apc-update/