RE: Calculating the Cost per Article in the Current Subscription Model

[Phil Davis <pmd8@cornell.edu>]

David Prosser brings up a good point -- that I did not factor in page
charges and other payments to publishers that do not flow directly from
the library.  He is also correct in that they are difficult to calculate
and account for.  Let's see if I can come about estimating these costs
from another direction ....

In our first OA report, I broke down the estimated per article costs in
more detail.  For Elsevier (where there are no page charges) the per
article costs in a producer-pays model would need to be about
$3,000/article for our institution to save money ($2,844 to be precise).  
Why is this figure much higher than the average article costs?  Simply
because we spend about 43% of our scholarly journal funds, yet only
publish 16% of our articles with this company.

If we remove all of the large commercial publishers from the model and are
left with the society, association, university press, small commercial,
government publishers, the cost per article would need to be less than
$400 for Cornell to save money ($393 to be precise).  Ok, now for the sake
of argument, lets assume that the full costs to publish an article are
$1,500.  We are looking at about $1,100 per article difference between
what the library has been paying and what the real costs would be.  This
shortfall is much higher than average society page charges (probably at
least double).  We need to conclude that page charges alone would not
cover the shortfall of funds for ARL institutions.

I realize that this is a round-about way at calculating whether page
charges and other author-subsidies are sufficient to change the outcome of
the model.  If anyone can come up with a direct way of doing this, please
let me know.  A society publisher would be able to itemize all funds
coming from a particular institutions and provide a case study.

--Phil Davis

At 06:24 PM 1/7/2005 -0500, you wrote:

> My concern is that the analysis takes no account of additional sources of
> funding for publication with the institutions.  Many authors at ARL
> institutions are paying significant amounts for publication in page
> charges, reprint charges, colour figure charges, etc.  However, the chart
> compares the average cost to the library in a subscription model (ignoring
> page charges, etc.) with the total cost per article in an open access
> model.  The comparison is not like-for-like.
>
> I have no idea what the average amount spent per article by authors is and
> it would be tedious to find out, but until we know I'm not sure how useful
> the chart is.  If it is tens of dollars per paper it makes little
> difference, but if 100s it does.  Until we know, we don't know!
>
> Also, the assumption behind the chart is that some institutions will be
> better or worse off.  However, if publication charges are seen as research
> charges and covered by the funding body then it surely does not affect the
> library so much.  There could be 'new' money coming into the system.
> There certainly is from the Wellcome Trust which has decided to cover
> publication charges for open access journals.  If more funding bodies do
> this then won't the cost/benefit question for individual institutions
> become moot?
>
> Best wishes
>
> David
>
> David C Prosser PhD
> Director, SPARC Europe
> E-mail:  david.prosser@bodley.ox.ac.uk