Re: Growth for STM publishers in 2008

[John Houghton <John.Houghton@vu.edu.au>]

Janice, et al.

While finding 'The stm Report' interesting and informative, in 
reference to work that I have been involved in I can't help 
noticing a repetition of the number of mistakes that were made by 
representatives of some in the publishing industry earlier this 
year and refuted at that time.

I refer to Section 4.9 (pp56-57) of The stm Report, which 
addresses the system-wide perspective on costs and cost savings. 
The text reads:

"A JISC report (Houghton et al. 2009) published the following 
year by the economist John
Houghton estimated system-wide savings accruing to open access 
publishing in the UK
alone at GBP 212m, less the author-side fees of GBP 172m, giving 
a net saving of GBP 41m. (This
appears roughly comparable in scale to the GBP 560m global 
savings estimated in the RIN
report.)  The largest single part of the savings (GBP 106m) came 
from research performance savings, including reduced time spent 
by researchers on search and discovery, seeking and obtaining 
permissions, faster peer review through greater access, and less 
time spent writing due to greater ease of access e.g. for 
reference checking. Funders should, according to Houghton, 
therefore be comfortable with diverting research funds to pay for 
open access charges because the savings in research performance 
etc. would outweigh the cost.

In response, publisher organisations (PA, ALPSP & stm 2009) have 
argued that the analysis was deeply flawed. It underestimated the 
efficiencies of the current subscription system and the levels of 
access enjoyed by UK researchers. Many of the savings 
hypothesized would depend on the rest of the world adopting 
author-pays or self-archiving models. The calculated savings 
would remain hypothetical unless translated into job losses; for 
example some 200 library job losses would be required to realize 
the estimated GBP 11m savings in library costs. Critics also 
argue that Houghton et al. underestimated the costs of switching 
to an author-pays model because they underestimated the true 
costs of publishing an article only, and because additional costs 
such as the infrastructure required to manage the many small 
publication charges were not included.

In addition to the system savings, Houghton suggested increased 
economic returns to UK public-sector R&D arising from increased 
access might be worth around GBP 170m. This appears speculative, 
resting on flawed and untested assumptions about the levels of 
current access and the marginal rate of return to any increased 
access."

  Just for the record:

     * The publishers' comments referred to (i.e. PA, ALPSP & stm 
2009) triggered a response from JISC which I note is not 
mentioned, but was sent to the publishers' associations and can 
be found at 
http://www.jisc.ac.uk/media/documents/publications/responseoneiaspmreport.pdf. 
Hence many of the mistakes (e.g. regarding job losses) could have 
been avoided rather than simply being repeated.

     * In response to publishers' comments, an addendum to the 
JISC report was issued which can be found at 
http://www.cfses.com/EI-ASPM/JISC%20EI-ASPM%20Report%20%28Addendum%20April%2009%29.pdf 
Its primary purpose was to further tease out the differences 
between models and between UK unilateral versus worldwide 
adoption of OA alternatives. Hence the issue embodied in the 
comment "Many of the savings hypothesized would depend on the 
rest of the world adopting author-pays or self-archiving models" 
has also been dealt with before and could have been avoided 
rather than simply being repeated.

     * The phrase "Critics also argue..." (e.g. page 57 of The stm 
Report and page 7 in the recent SME report also by Mark Ware (see 
below)) fails to carry any references or indicate who the critics 
are. Are there such sources?

     * Page 57 of The stm Report states: "... because additional 
costs such as the infrastructure required to manage the many 
small publication charges were not included." This claim about 
the cost of author-pays payment management is also repeated and 
is incorrect. A cost for author-side payments was included in the 
model.

     * The stm Report states "... underestimated the efficiencies 
of the current subscription system..."  In fact, the returns to 
R&D aspect of the analysis is based on introducing accessibility 
and efficiency into a standard Solow-Swan model as negative or 
friction variables and looking at the impact of reducing the 
friction. Hence, "the efficiencies of the current system" are the 
baseline... they are not underestimated, overestimated or 
estimated in any way at all, they are taken as given.

     * The stm Report states: "... speculative, resting on flawed 
and untested assumptions about the levels of current access...". 
The levels of access are discussed at length in the JISC report, 
as is the basis for the parameters used in estimating the 
potential impacts on returns to R&D spending. Data sources and 
references are given. Moreover, its difficult to see how the 
potential 5% increase in accessibility modelled in the JISC study 
could realistically be described as "underestimated... the levels 
of access enjoyed by UK researchers" in the light of the 
evidence. Just to take one example, a recent survey of UK small 
firm (SME) access to journal articles by Mark Ware Consulting 
(http://www.publishingresearch.net/SMEaccess.htm) found that 73% 
of UK-based SMEs report difficulties accessing the journal 
articles they need, and that just 2% of SME, 7% of large firm and 
17% of higher education-based researchers reported having access 
to all the articles they need for their work (page 13, table 2). 
The same report notes that there are 4.7 million businesses in 
the UK of which 99.3% have fewer than 50 employees, and it would 
appear from reported sample sizes that 2% of SMEs equates to just 
4 firms. On page 22 the report notes that 71% of SMEs reported 
using open access journals and 42% reported using institutional 
repositories. On page 30 the report also notes "Several firms 
were enjoying access via the libraries of the universities where 
they had previously worked. It was not entirely clear whether 
this use would have been legitimate under the terms of the 
libraries' licences." Only Mark knows whether there was any 
overlap between the 4 SMEs and the "several firms...", or between 
the 4 SMEs that reported having access to all the articles they 
needed and the 132 small firms that reported using OA journals.

Regards,
John Houghton

Victoria University