No, Mandating Self-Archiving Is Not Like Invading Iraq! Part II

[Stevan Harnad <harnad@ecs.soton.ac.uk>]

On Sun, 24 Dec 2006, Sandy Thatcher wrote:

> And what about copyediting? You nowhere mention this as a cost, 
> and it can be significant. In my experience, very few academic 
> editors are able to do line editing very well (nor should they 
> spend their valuable time doing so anyway), and very little of 
> academic writing is not in need of editing. (I understand that 
> the British have a different attitude about copyediting, but in 
> the U.S. it is generally valued a lot, and expected, by most 
> authors.) I speak from experience here, as I did copyediting 
> full-time for the first three years of my publishing career and 
> continued it part-time for another twenty years. If you abandon 
> copyediting, you will have a significantly degraded product. 
> Good copyediting comes at a cost, though, at about $25 an hour.

It is an empirical question whether copy-editing will be part of 
the downsized OA publishing essentials. (The one substantive 
finding of the PRC study was was that librarians don't seem to 
set much store by copy-editing in their acquisition/cancellation 
criteria.) To the extent that copy-editing continues to be part 
of the downsized OA publishing essentials, it will be paid for, 
on the OA publishing cost-recovery model, out of the 
institutional subscription savings, per outgoing article 
published. (How many hours of copy-editing do you estimate an 
article needs, at $25 per hour, in the online age?)

> Yes, author fees can cover this cost, too, but your model for 
> transferring costs from libraries to on-campus editorial 
> offices or BMC-type publishers assumes a smooth transfer. Have 
> you had any experience in university administration? Nothing 
> works that smoothly in universities, I assure you. A one-to-one 
> transfer of library serials expenditures to faculty publishing 
> fees is no simple matter, nor is there any guarantee that the 
> funds freed up by cancelled subscriptions would migrate 
> directly to author fees anyway. There are plenty of other uses 
> to which such funds might be put. Libraries have multiple 
> needs, and supporting faculty publication may not immediately 
> be at the top of their lists. Even today, when costs might be 
> seemingly passed on easily to faculty who avail themselves of 
> library e-reserve operations, it doesn't happen because the 
> administrative costs of such transfers are perceived by some 
> libraries as steeper than the costs of paying for all e-reserve 
> permission fees themselves.

On your own hypothesis of sudden change (H3), two things will 
happen, suddenly, and at exactly the same time: sudden collapse 
of institutional subscriptions and sudden institutional windfall 
savings on subscriptions. Necessity is the mother of invention; 
with journals suddenly unable to make ends meet, and obliged to 
charge for publication on the OA model, and institutional authors 
suddenly obliged to pay to publish, it will not be lost on anyone 
that authors' own institutions have the sudden windfall resources 
from which to pay for their sudden rainfall of costs.

(I think you are perhaps getting carried away with free 
speculation if you feel that it is part of the "worst case 
scenario" that universities are simply too thick to see that they 
must transfer some of their windfall subscription savings to 
their authors' sudden rainfall of OA publishing needs under these 
acute conditions. Such oversights are among the pitfalls of 
unconstrained speculation!)

> Your model also assumes that subscription savings will balance 
> out author fees at any given university. That is a very big 
> assumption to make. Yes, the most active authors are probably 
> at the most research-intensive universities, but I doubt there 
> is any one-to-one correspondence. Some universities may find 
> that they have to spend much more in author fees than they save 
> in subscriptions, whereas others may find the reverse. Also, I 
> suspect that, to the extent this correspondence exists in 
> science, it exists much less so in the humanities and social 
> sciences. Over time we have found in university press 
> publishing that there has been a very significant dispersion of 
> talent to non-ARL campuses, such that we are publishing many 
> more authors from second- and third-tier universities and 
> colleges than we did, say, twenty or thirty years ago.  The 
> savings from subscription cancellations on those campuses may 
> well not come close to covering author fees for their faculty, 
> who will thereby be disadvantaged in getting their writing 
> published unless their universities can tap some other source 
> of revenue for that purpose.

(On this question of net provider vs. net consumer institutions, 
both for subscriber journal input and for published article 
output, please do the google search:

     site:www.ecs.soton.ac.uk amsci net providers consumers

discussion already began in 1999!)

There is every reason to expect that savings and costs will 
balance out comfortably. The two main reasons are these:

(1) There is undoubtedly a high correlation between the size of 
an institution's research output and the size of an institution's 
research journal intake.

(2) Downsizing means that substantially less money will be 
changing hands (costs will probably be less than a third of 
today's expenditure). This means there is a 2:1 buffer against 
any net-journal-consumer net-article-provider imbalances).

> Now, you might say, OA journals will take these inequities into 
> account and charge lower fees to such authors, or waive them 
> altogether. But then you introduce a whole new level of 
> administrative cost into the system because there has to be 
> some way to verify "hardship" cases, especially if you are 
> dealing with authors in this country and not from some very 
> poor developing countries. If all this is done on the "honor" 
> system, you open the system to a significant level of 
> corruption and free-riding. Moreover, the costs still have to 
> be paid, and this scenario would mean that the wealthier 
> universities would again be supporting the cost of the whole 
> system that benefits everyone, as they do now for university 
> presses.

The research-active institutions will pay the lion's share of the 
research publication costs because they do the lion's share of 
the research. They also pay the lion's share of the subscriptions 
today, and hence will have the lion's share of the savings out of 
which to pay for OA publishing. The problem of authors from poor 
institutions and the problem of unaffiliated authors is a small 
minority problem in the OA world, and it will be solved, in very 
obvious ways. In any case, it is not part of the sudden-change 
scenario (H3) that we are contemplating here, which was supposed 
to be about the hypothetical hardships of *publishers* struggling 
to make ends meet after the hypothetical sudden collapse of 
subscriptions, not about the hypothetical hardships of *authors* 
from poor institutions, struggling to pay OA publishing costs.

(No publisher is worrying about the hardships of all those 
researchers worldwide who cannot afford access today; OA is the 
remedy for those researchers. Let publishers not now -- as a 
pretext for continuing to deny researchers that remedy -- plead 
that it is in order to protect those same researchers as authors 
that the access is to continue to be denied! That would be the 
tail wagging -- or rather restraining -- the dog, and the tune 
would not ring true!)

> You're also assuming that library fees would be readily 
> transferred outside the university to publishers like BMC or 
> society OA publishers through author fees. That would introduce 
> yet another level of complexity into the system, as I do not 
> share your assumption that universities would as readily allow 
> transfer of funds to such an entity as they would to another 
> university-based publisher.

Sandy, I think you already said this once above -- that 
universities will not redirect incoming subscription savings to 
outgoing publication costs: The answer is that of course it is 
not "complex" and of course they will. (And they will pay the 
author's chosen journal; it is irrelevant who the publisher is -- 
commercial, learned-society, university, independent non-profit, 
or mom-pop: Relevant only is the author's choice of journal.)

End Part II. Part III follows.

Stevan Harnad