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

[Phil Davis <pmd8@cornell.edu>]

The goal of this model was for readers to challenge and question its
assumptions and the model itself, and David and Heather are doing just
that.  Because I'm not a publisher, I will leave the cost per article in a
producer-pays model up to those who know much more than me.  I will answer
David's other questions about the model:

4. Our taskforce assumes that 50% of an ARL serials budget goes to
purchase scholarly journals.  The rest goes to purchase indexes and
databases, magazines, newspapers, encyclopedias, and anything that can be
considered a regular subscription cost.  It is really difficult to take a
line item budget and start dividing it up this way.  We tried it at
Cornell and came up with around 50% and assume that other ARL libraries
are about the same.

ISI indexing coverage

I'm a bit surprised that no one to date has challenged the percentage of
scholarly articles that are indexed by ISI.  Based on an idealized model
in which ISI only indexes the most prolific journals (in rank order),
their 8,769 indexed journals index only 44% of scholarly journals, but an
amazing 92% of the articles published!  We all anecdotally know this
figure to be much too high, and I haven't received anything more accurate
from the staff at ISI.  Lowering this indexing percentage down to say 70%
has an amazing affect on the results -- much more than adjusting the cost
per article range.

Your concluding paragraph is very cogent and needs no paraphrasing except
some additional words on the implications if this economic model is
correct.  Firstly, this very conservative model predicts that most ARL
institutions would pay more in a producer-pays model than in a
subscription model.  This is consistent with anticipated results: in a
producer-pays model, those who produce more will pay more.  Researchers at
the top 100 research institutions produce the vast majority of published
research in the US and Canada, and if the full costs of publishing are
concentrated among these top producers, it is reasonable that they will
pay more.  Not all ARL institutions are predicted to pay more -- there are
a handful of smaller institutions with very large library endowments
(which is why they are part of ARL), who may pay less.

Whether or not a producer-pays Open Access model is cheaper overall, some
ARL institutions would pay considerably more in a producer-pays model and
the real challenge will to find ways to reallocate or redistribute money
to make this happen.

Sincerely,
Phil Davis

At 12:51 AM 1/4/2005 -0500, you wrote:
> I'd like to mention some factors I find helpful in interpreting this data, 
> as also noted by
> Heather and by Sally.
>
> 1. Phil, as I interpret footnote 4, "figures include overhead and
> profit"--please correct me if I'm not seing this right, but the
> publisher's cost including overhead and profit is the price the publisher
> charges the library. As librarians, you and I think of that as _the_
> cost, because it's our cost. (subject to my note 3)
>
> It is however _not_ the cost to the publisher.  The cost to the publisher
> does not include profit; he can price his product to attain whatever
> level of profit he thinks is appropriate and possible. The cost for the
> publisher to produce a journal , including overhead, might be $1000, and
> if he prices it at $2000 then the library's cost is $2000, and the
> publisher makes a very good profit. . With the same $1000 cost to the
> publisher, he might price it at $1000 and break even, which may be all a
> particular non-profit publisher wants.
>
> 2. Even considering the cost to the publisher, it represents his current
> costs, and the price as based on that cost. It does not represent what he
> might be able to sell it at if he produces more efficiently, produces
> more cheaply, has lower overhead, or - even -- gets more subscriptions!  
> If any of these happens, he often chooses to produce larger journals, but
> might choose higher profits, or lower prices.
>
> 3. I think some but not all of the library serials budget figures include
> some acquisitions costs, such as fees for subscription agents and
> postage. These will not be major components, but they do represent
> additional costs to the libraries. There is also library overhead, which
> is I think never included in these figures--not even the salaries of the
> serials staff.
>
> 4. The 50% you use for the percent due to scholarly journals, does this
> also include indexing and abstracting services?  I think libraries do
> this differently, and it may be a major component.
>
> 5. The data is changing--figures from price lists, library statistics,
> annual reports, and so on are generally out of date when published, but
> we cannot hope for better.  Experience helps in knowing how to adjust
> them.
>
> Concluding, your data about shift to a producer-paid model assumes equal
> values for all of these factors under both models.  I agree that it is
> the right way to think about it.. It is not prudent to plan on the
> assumption libraries could force publishers to lower overhead or decrease
> profits. Even if we could, that fwould apply to either model.  As almost
> all of us recognize, the rationale for a change to OA journals is not to
> reduce the cost, but to improve the access. It's the economic effect of
> that change on individual universities that your spreadsheet shows, and
> it will obviously be necessary to compensate for the change in some
> manner.
>
> Yours,
>
> Dr. David Goodman
> Associate Professor
> Palmer School of Library and Information Science
> Long Island University
> dgoodman@liu.edu