Re: Costs of open access publishing - the Wellcome Trust

["Fytton Rowland" <J.F.Rowland@lboro.ac.uk>]

Rick Anerson wrote:

> I'm reminded of LaRouchefoucauld's observation: "There is nothing more
> horrible than the murder of a beautiful theory by a brutal gang of
> facts."  The fact that OA providers are starting to talk more and more
> like commercial publishers ("Sorry it's so expensive, but quality always
> is") should tell us something about the actual costs of scholarly
> publication in the real world, as opposed to the theoretical costs of
> scholarly publication in an imaginary utopia.
>
> Here's what I think is the real-world situation right now:  Yes, some
> journal publishers have been charging prices that are unjustifiably
> high.  No, open access models cannot be provided nearly as cheaply as OA
> proponents have suggested.  Therefore, here's my prediction for the
> future: As OA models continue to evolve, we can expect to see the price
> of OA continue to climb until it reaches equilibrium.  That point will
> probably be somewhere below the cost of the average Elsevier journal.
> It will not, however, be nearly as low as OA advocates think (even now).

There are two elements of the charges of OA journals.  One is due to their
selectivity.  If their model is based on a publication charge only, then
the higher their rejection trrate, the higher that charge will be, since
the published papers have to carry the refereeing costs of the rejected
ones.  A journal that accepts only 20% of submissions will have to have a
charge four times as high as a journals that accepts 80% of submissions.  
This can be seen as "the cost of quality", though with two caveats.  
Top-quality journals could have a submission charge instead of (or in
addition to) a publication charge, thus recovering some costs from the
rejected papers.  And rejection rates vary between disciplines: in the
hard sciences an 80% acceptance rate does not necessarily imply a poor
quality journal.

The other element is dependent on what the journal actually does (in
addition to refereeing). Established publishers tend to assume that the
various elements of "value-added" that they put into their journals are
all wanted, and valued, by one or both of their two markets (authors and
readers).  This may not be so.  These features may be adding to costs
without really being wanted by the users.  If so, a new entrant to the
scholarly publishing industry may be able to have a much lower cost base
by simply offering a "bare-bones" publishing service without the (perhaps
unwanted) bells and whistles.

Fytton Rowland, Loughborough University, UK.