On the left, we see the $8 billion U
.S. in revenue from journals received by the scholarly publishing industry. Of this, about $5.6 billion is revenue from academic library subscriptions. Compare this to the $2.5 billion it would cost to publish every one of the world's estimated 1.5 million scholarly articles produced annually on a worldwide basis at the BioMedCe
ntral standard fee of $1,680, for a total of $2.5 billion. This would be full global open access to scholarly articles at less than half of what academic libraries are contributing to the system now. The savings would be even greater at average PLoS ONE fee of $1,350. On the right hand side, we see what would if happen the current $188 average per-article revenue for journals using OJS were to apply across the board, the global total of $.2 billion would be a tiny fraction of the current spend - less than 3%.The next chart, Comparison of per-article revenue, illustrates what to me is a central point is considering the economics of transition to open access, that is: as we shift the economics from purchase to production, the cost per article is critical in determining the feasibility and sustainability of the system. On the left, we see the average per-article revenue received in the current system of over $5,000. This is compared with the BioMedCentral standard article processing fee and PLoS ONE, about a third of the current per-article revenue. Since both BMC and PLoS ONE are doing well financially, this illustrates that high quality publishing is possible at less than a third of the present revenue. On the right hand side, we see the average revenue for an OJS journal of $188, illustrating that the traditional scholarly gift economy can manage essentially the same work for about 3 percent of the current average revenue.
These numbers are meant to be suggestive only, one guideline to consider in the transition to open access.
Recommendations
If the goal is an affordable open access scholarly publishing system, it makes sense to support scholar-publishers of the type included in the OJS survey, and it makes sense to support cost-effective charges like the BMC standard and PLoS ONE.
Sources
The $8 billion U.S. annually for scholarly journals revenue is from the Research Information Network (2008)Activities, costs and funding flows in the scholarly communications system in the UK, retrieved from http://is.gd/3Q7cm as cited in the STM report. 2009. International Association of Scientific, Technical and Medical Publishers (STM). Retrieved from: http://www.stm-assoc.org/
Global annual scholarly peer-reviewed articles is estimated at 1.5 million per year, as reported by Björk et al, Global annual volume of peer reviewed scholarly articles and the share available via different open access options, ELPUB 2008, retrieved from http://www.utsc.utoronto.ca/~elpub2008/presentations.html
The PLoS ONE and BMC article processing fees are from the respective websites as of March 1, 2011.
The $188 average revenue per OJS journal is from a survey by Edgar, Brian & John Willinsky. A Survey of the Scholarly Journals Using Open Journal Systems. 2011. In press. Retrieved from http://pkp.sfu.ca/node/2773
This post is an update and elaboration of earlier charts I developed for IJPE and various presentations.
Typo correction March 5: number of peer-reviewed articles per year is 1.5 million, not 1.5 billion. Thanks to Douglas Carnall for spotting the error.
