From: Paul Courant <[log in to unmask]>
Date: Tue, 27 Jan 2015 08:01:47 -0500

Joe, you state the following:

So the question could better be put this way:  What is the retail
price for all the subscription journals (that are
available in electronic form), and how much would a customer actually
have to pay for them?

I can imagine discovering the retail price for all of the subscription
journals that are available in electronic form.  It would be hard to
do, for the reasons that people on this list have adduced.  In
thousands (tens of thousands?) of cases the prices aren’t posted, and
the ingenious sampling idea that Kathleen Folger and others on the
list have proposed suffers from the possibility that journals with
posted prices differ in some systematic way from journals without
them.

But your second question simply cannot be answered in any general way
via any imaginable research project.  It is well-documented that
journal publishers engage in price discrimination and the range of
prices for an institutional subscriptions is large.  (Ted Bergstrom
and a number of others, including me, have  a paper on this subject in
the Proceedings of the National Academy of Sciences.  The paper is
behind a pay well, but is available free from Ted’s website.  Go to
http://www.econ.ucsb.edu/~tedb/Journals/jpricing.html and scroll down
until you see a link that says “here is a link to the paper.”)
Michigan pays different prices than Minnesota and both pay different
and much higher prices than institutions that are farther down the
Carnegie classification scheme.  So you have to identify the customer
when you ask what "a customer would actually have to pay," and there
is no way around this.  There simply is no well-defined market price
of many journals.

Paul

----
Paul N. Courant
Harold T. Shapiro Professor of Public Policy
Arthur F. Thurnau Professor, Professor of Economics, and Professor of
Information
The University of Michigan