Re: The True Cost of the Essentials (Implementing Peer Review)

From: Albert Henderson <>
Date: Tue, 8 Jan 2002 02:05:59 +0000

On 19 Dec 2001 "Arthur P. Smith" <apsmith_at_APS.ORG> wrote:

> On Mon, 17 Dec 2001, Albert Henderson wrote:

> > on Fri, 14 Dec 2001 Stevan Harnad <> wrote:
> >
> > > "4. Whereas all refereed research should be fully accessible
> > > on-line without cost to all would-be users worldwide, it is
> > > nevertheless not altogether costless to produce. The main change is
> > > that dissemination and archiving cost incomparably less on-line
> > > than on-paper and hence the on-line dissemination/archiving costs
> > > per article effectively shrink to zero.
> > >
> >
> > You can claim to save only 9 cents per article with
> > online distribution!

> 9 cents per article per subscription. For a journal with 2000 print
> copies produced, that's $180 per article. For a journal producing
> and selling only one print copy, yes 9 cents would be your savings.
> At least that's the only way one can possibly understand the numbers in:

Not true. The publisher must treat the 9 cents as a variable cost,
rising or falling with the numbers of subscribers. As such, the
variable is of little concern, even if increased numbers of articles
force the total price upward. Saving the variable runoff cost, as
claimed, is particularly laughable because it comes at the cost of
expensive infrastructure and shifts of production (paper, energy) to
the reader.

> > King, McDonald and Roder estimated the pre-Internet
> > costs of U S science journals. They put per-article
> > prerun costs at $1050 in 1977; runoff costs were
> > 1981. p. 218-219]

> It does matter what factors are being included in quoted numbers!
> Per article in the recent discussion really meant per article,
> not per article per subscription, or "price per page" to the library,
> as is often quoted.

You are missing the point.

Members of the policy community, particularly those controlling library
budgets, bear a major responsibility for the rises in subscription
prices related to nonvariable costs of production.

The publisher faces a fixed cost ($1050 per article in 1977) that rises
'per-subscriber' when the numbers of subscribers decrease (as they have
for 30 years). Each subscriber must contribute more to support this
fixed cost as a result of subscription cancellations. It means that the
'price per page' rises.

Moreover, the science policy that constantly increases R&D spending
should acknowledge its major role in total library subscription costs.
Each library subscriber must pay more as the numbers of articles
increase, about 5 per cent each year (as they have for 336 years). The
fact that libraries are unable to meet this challenge testifies to a
gaping hole in the sincerity of policy insiders.

Thank you for your comment.

Albert Henderson
Received on Tue Jan 08 2002 - 02:06:58 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:46:22 GMT