On 24 October Dr Klaus Graf kindly posted some links in response to my post, ‘Open access publishing revisited’ (13 Oct). So far I have responded to one of the articles that Dr Graf provided links to. Here I’d like to respond to a second, namely ‘Full open access to articles – with library savings of over 70%’ by Heather Morrison in The Imaginary Journal of Poetic Economics:
Morrison’s argument is that if libraries were to fund open access publishing for the 1.5 million peer-reviewed articles published per annum, this would produce huge savings – the cost of subsidy would be much less than the current cost of subscriptions.
Morrison presents her argument as a straightforward equation: the average cost per article (as indicated by the proxies she uses) multiplied by the number of articles gives us the total cost of publishing peer-reviewed articles, which is less than the cost of library subscriptions (estimated at 70% of total subscriptions). Morrison states her conclusion as follows: ‘Calculate the ratio and voila! Libraries CAN have our cake and eat it too – full open access with cost savings’.
There is a good deal of estimation involved (Morrison herself calls the article ‘broad-brush’). Given her aim – ‘to view how achievable open access is from an economic standpoint’ – we should examine the argument rigorously.
One potential challenge arises from a point I made before in response to one of the articles that Dr Graf provided links to (see ‘Open Access: response (I)’, 3 November): if publication is fully subsidised, the number of articles is likely to grow (notwithstanding the fact that they are peer-reviewed). That, of course, may be seen as a good thing – but it would add to the total cost.
Another challenge arises from the problem of free-riders. Morrison writes ‘libraries worldwide could fund full open access’: but how would one ensure that all libraries paid their fair share? And how would one calculate fair shares? Such questions are central to the question of achievability.
A third challenge arises from the question of whether the proxies Morrison uses are reliable. These proxies (to which she helpfully provides links) are taken from:
- BioMed Central: see http://www.biomedcentral.com/info/authors/apcfaq#flatcharge
- PLoS One: see http://www.plosone.org/static/policies.action#pubfee
As a book publisher, I don’t have a feel for whether the article costs she uses are typical. As Morrison herself notes ‘the average cost per article matters. To keep things simple, this macroanalysis only considers one business model for open access, and only two publishers’. However, I do note that BioMed Central claims that ‘BioMed Central’s article processing charges are extremely competitive. In general, BioMed Central’s charges are amongst the lowest of any publisher offering open access publication’ and supports this claim with a comparison table.
The PLoS One website says that ‘To provide open access, PLoS journals use a business model in which our expenses—including those of peer review, journal production, and online hosting and archiving—are recovered in part by charging a publication fee to the authors or research sponsors for each article they publish’.
That ‘in part’ is important. It looks as though Morrison is taking as the full cost of publishing an article a figure that in fact represents only a part of the cost. (On this note, I should add that BioMed Central says that it ‘defrays’ its costs. Presumably this does indeed mean, as Morrison assumes, ‘covers in full’ – though if not, that would constitute a difficulty.)
It is, of course, entirely reasonable for Morrison to adopt a ‘broad-brush’ approach and to make some simplifying assumptions. How else to provide a concise overview of the argument? And it is admirable that she provides links to her sources. But, perhaps unavoidably, the consequence of such an approach is, it seems, a certain lack of rigour. Her ‘voila’ smacks too much of the magician plucking a rabbit out of a hat.
However, the big question lies, I think, elsewhere. Morrison argument is that the total cost of publishing amounts to (or could amount to) less than the total cost of subscriptions. My question, then, is: what is being lost – or what are publishers spending money on that they don’t need to?
To put this question in context: the conclusion of my original posts on open access publishing (1, 3, & 7 Sept), namely that ‘there is no such thing as a free monograph or journal’, was based on the view that the open access model primarily provides a way of transferring cost, rather than reducing it. The above question is, therefore, fundamental.