[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Decision making by Libraries on serials and monographs and useage (re puzzled by self-archiving thread)

Joe, I think Anderson's primary point is about market extension (number of titles for which there are customers), more than intensification (number of copies for which there are customers) - although both are important.

My point on possible differences between average and marginal returns was based on reported citation differences between subscription and OA articles, particularly in so far as it related to research use (intensification?), and on the "long tail" idea in general, particularly in so far as it related to potential non-research use (extension?) - although the mapping was obviously far from perfect.

IF Amazon sells anything like twice as many titles as are stocked in the largest physical bookstore, it does at least suggest that extension might add significantly to intensification. And, to get back to the original point, rather than speculating as to whether marginal returns are decreasing or possibly even increasing, we used average returns as a starting point for preliminary estimations.

Perhaps its an Australian (small/remote country) perspective, but tails seem an important potential benefit of OA from here. Try getting a history of Fiji published by one of the major North American or European publishers and, regardless of its scholarly merit, you would be unlikely to succeed, because they don't think the market for it (that they serve) is big enough. And yet, if you look at download statistics from institutional repositories at local universities where there are significant Asia-Pacific research schools, you see thousands of downloads of things like histories of Fiji (a number of downloads sometimes 2, 3 and 4 times the average print run of the popular/mainstream academic titles that are published). So there does appear to be a substantial un(der)-served market - for more titles and more copies.

Regards, from the tail end of the world...
John Houghton

Centre for Strategic Economic Studies (CSES), Victoria University
VoIP: (FireFly) 88207699 (Skype) John.Houghton
E-mail: john.houghton@pobox.com
Web: www.pobox.com/~houghton


Joseph J. Esposito wrote:

I don't want to engage the argument between Sally Morris and John Houghton, but I do want to point out an error of fact in one of Houghton's sources. Houghton quotes Chris Anderson's The Long Tail (not a very good book, by the way: read it and see) to this effect: " 'if the Amazon statistics are any guide, the market for books that are not even sold in the average bookstore is larger than the market for those that are.'" This is simply wrong.

Amazon's statistics are not a guide. It is apparently true that Amazon's sales come predominantly from "long tail" titles, but Amazon has enormous market share for those titles--for some titles that share is 100%. For better-selling titles (the 130,000 Houghton cites, though a figure a third of that would make more sense, if the aim is to reflect the realities of bricks-and-mortar bookselling) the share is distributed across thousands of booksellers. The market for books outside the top 130,000 is decidedly not bigger than that for the 130,000.

To the extent that Houghton's argument is propped up by Anderson's authority, it has to be said that Houghton's thesis is unproven.

Joe Esposito

----- Original Message -----
From: "John Houghton" <John.Houghton@vu.edu.au>
To: <liblicense-l@lists.yale.edu>
Sent: Thursday, January 25, 2007 4:15 PM
Subject: Re: Decision making by Libraries on serials and monographs and
useage (re puzzled by self-archiving thread)

Sally Morris wrote:
I am no economist so my questions are common-sense ones (I think)

Increasing access I can understand in principle - but how does one increase 'efficiency'(as an input)? Your most prominent definitions of 'efficiency' are related to 'relevance' and I really don't see how that could be increased. Wouldn't the arguments be more convincing if one looked at the increase in just one variable, anyway?
As noted before, efficiency is used in two related senses: the usefulness/use of the knowledge created by R&D and the efficiency of the conduct of R&D. In the report (pp31-34 and Appendix II) we outline some of the potential impacts of enhanced access, including a range of ways in which the efficiency of research might be increased (e.g. increased speed of discovery, reduction of duplicative research, etc.) and its use might be extended (e.g. enhanced access to industry, government and society, the emergence of new industries such as weather derivatives, etc.). These are discussed in the context of developing an "impacts framework" that focuses on the issues of access, use and efficiency. As for treating access and efficiency separately: its an option, but we thought that access, use and efficiency of R&D were likely to positively be related... for all the reasons outlined in the literature review in Appendix II.

As to the one-to-one relationship between a given percentage of increased access (or anything else) and increased benefit - could you clarify that? I'm not assuming that most users have access already - just that those who do are likely to be those most able to benefit, and that ability to benefit will decline as access increases. The same would go for any impact on the efficiency of the users' own research.

To a simple non-economist like me, it all seems to rest on huge and rather implausible assumptions...
I'm not sure we are making any assumptions. The estimates are presented in the form: IF... THEN... Obviously, the things following the IF are the variables. We are simply putting forward range estimates of the possible impacts on social returns to R&D, and based on a literature review we use plausible ranges of social returns from 25% to 75% and 1% to 10% increases in access and efficiency (the thinking behind which we discussed in the last message). Purely for the purposes of discussion we then use examples based on a 25% rate of return to R&D and 5% increase in access and efficiency.

As noted before, we use average rates of return because we are not changing the level of R&D expenditure. The extent to which there may be diminishing marginal returns to access depends on how far we are from optimal access at the moment. In his discussion of the "long tail", Anderson noted that:

/"What's really amazing about the Long Tail is the sheer size of it... Take books: The average Barnes & Noble carries 130,000 titles. Yet more than half of Amazon's book sales come from //outside// its top 130,000 titles. [so].. if the Amazon statistics are any guide, the market for books that are not even sold in the average bookstore is larger than the market for those that are. In other words, the potential book market may be twice as big as it appears to be, if only we can get over the economics of scarcity./"

As also noted before, the evidence from the Open Citation Project and from OA repository download statistics suggests that there is much more reading/use with OA... perhaps, when one adds zero pricing (to the user), there may be a scholarly publishing long tail. If the market that isn't reached is bigger than that which is, and its been reported that OA articles get 2-5 times the citations, suggesting that use by researchers alone may increase by 100% to 400%, then the difference between average and marginal returns is unlikely to be large - be they increasing or diminishing.

Regards,
John Houghton

--
Centre for Strategic Economic Studies (CSES), Victoria University
VoIP: (FireFly) 88207699 (Skype) John.Houghton
E-mail: john.houghton@pobox.com
Web: www.pobox.com/~houghton