RE: Price discrimination for academic subscriptions (discussion)

["David Goodman" <David.Goodman@liu.edu>]

It should not be overlooked that the practice of charging flat rates is
also discriminatory pricing: it discriminates strongly in favor of the
largest schools. (This was less true in the print era when the largest
schools would hold multiple print subscriptions to important journals,
thus effectively producing a curve.)

Use-based pricing also needs a curve of some sort: most schools would be
much more willing to pay $35/article for Biochim Biophys Acta, where they
might need 50 articles a year, than they would for Cell, where they might
need 500, to mention two journals from the same publisher. Even for the
same journal, a school without a ecology department might need 10 articles
a year from Oecologia (price about $4000/yr--different publisher, btw),
and be willing to pay $35 each; a school with a large department might
need 1000, and be less willing. (All numbers are obviously merely
illustrative, but no likely variation of them will lead to a different
conclusion.)

There are two hidden factors that greatly affect the impact of any 
per-use scheme.

First, once an institution has purchased an article, is there an
additional charge for other users of the article from the same
institution? If there is, it greatly increases the cost for the larger
institutions, if not, it favors them.

Second, is there an additional charge if the same individual user accesses
an article a second time? If so, it provides a strong incentive for users
to store everything they have read on a local hard disk, rather than by
saving the links, which may not be the most efficient way of working. If
not, it requires tracking the individual users, which most of us would
find totally unacceptable. (An alternative which might possibly be
workable is a complicated retrieval code system). I note as a further
complication that pricing differently depending upon whether an article,
is read, printed, or downloaded is now technicaly feasible.

One can obviously have various combination schemes, typically involving a
ceiling price, often the traditional subscription price. All of these
yield the same problems as now, that of deterining the appropriate price
curve.

In any case, any model of per-use pricing eliminates the possibility for a
research institution to own and archive the complete contents of a
journal.

I do not think there can be any equitable business model for the typical
STM journal in its present form. For the user, only open access of some
sort is equitable. Models such as PLOS or BMC meet this criterion, but
then raise problems for authors of different abilities to pay, though the
questions of equity may prove less intractable at the producer end.  I
conclude the only truly equitable model to be a open archives proposal
similar to the arXiv model, or the original Harnad or Varmus proposals.

Studying or debating which scheme of charging extortionate prices is the 
least unfair is not a productive approach. 

Dr. David Goodman
Associate Professor
Palmer School of Library and Information Science
Long Island University
dgoodman@liu.edu