Re: Building collections in a bad economy

[Greg Raschke <greg_raschke@ncsu.edu>]

Joe,

With the economic situation moving libraries to innovate more 
quickly than they might have, I think on demand is coming for 
books faster than you imply below.  NC State, Texas, and others 
already do this for a portion of their e-book collecting. 
Vermont has done interesting work with on-demand purchasing for 
print collections.  We are looking to implement on demand for a 
significant portion of our book collecting over the next two 
years.

Prospective buying at academic, college, and research libraries 
consistently results in 30-40% of titles that do not circulate 
(with significant variation by subject, but tending to hold true 
across a large enough set of books) in a 5-10 year period, no 
matter the selection method.  Some might argue this is 
acceptable, particularly for research libraries, but the economic 
climate and improving technologies make that argument much 
tougher to make.  Given the technologies available to us to 
deliver e-content at the point of request (or 2nd or 3rd request 
to initiate a purchase) and delivery print within 2-3 days (or 
faster with pod), that is an unacceptable non-use rate for 
prospective purchasing.

In short, you are absolutely right that it will better align our 
collecting and expenditures to use and I agree for e-content, we 
will not see savings.  For print purchasing though, I think 
libraries will realize savings that they can either invest in 
more e-content or use to offset reductions.  You can load the 
records for what would have come on an approval or slip plan and 
then let users decide when they want it.  The other advantage of 
such a system is that it will help overcome the delayed e-book 
availability problem since you do not necessarily purchase the 
print hardcover when it first comes out, only when the first user 
initially needs it and with the choice of formats rather than e 
trailing p.

This is problematic for publishers and is a significant issue for 
university presses and other smaller publishers (for which I am 
very concerned, despite what this post might imply), but I think 
it is a likely direction for many college and academic libraries. 
I foresee a system where we centrally purchase some large e-book 
packages, purchase titles we know will move such as computer 
programming works, move the rest to e or p purchase on demand, 
and fill in gaps with a much smaller amount of title by title 
selecting.

That is a long-winded way of saying I think you are right, and 
libraries, publishers, and vendors need to start scoping out 
these collecting and business models now.

Greg Raschke
Associate Director for Collections and Scholarly Communication
NCSU Libraries
919-515-7188
greg_raschke@ncsu.edu



Joseph Esposito wrote:

> I have been pondering several of the public statements about 
> collection-building in these very difficult times and have been 
> struck by the fact that there is little discussion of testing 
> on-demand services.  I say "testing" advisedly, as a 
> comprehensive on-demand system is bound to throw up 
> unanticipated items, and it is best to solve a problem with a 
> small set of elements rather than a large one.
>
> By "on-demand" I refer to the practice of only purchasing 
> something when it is actually used.  This is not the norm with 
> libary publications, of course, where most publishers sell 
> things that include many components that are literally never 
> used.  I know there is a case to be made for collecting 
> everything even if some parts are never used, but it is a case 
> I personally find it hard to make.
>
> There are at least three large problems with establising an 
> on-demand service; and if we spend a few minutes on this 
> problem, we will find three more.  The first is simply the 
> systems issues, which cannot be underestimated.  To sell things 
> by use requires reliable systems that enable proper feedback, 
> without endangering the privacy of individual users.  This 
> means publishers would have to "look" into libraries' internal 
> systems.  The second is the back-end production process on the 
> publishers' parts.  When we say "on-demand," at what point in 
> the process is the demand created?  Print on demand? 
> Digitization on demand? And what the heck, how about authoring 
> on demand?
>
> And then there is the arena of endless argument, the third 
> point, which is pricing.  For an on-demand system to work, 
> individual units would cost more, perhaps MUCH more, than their 
> pro rata share of collections.  Think of the world of the audio 
> CD.  A CD costs $10 for ten songs.  Therefore there is an 
> erroneous expectation (call this The iTunes Fallacy) that each 
> song should cost $1.  When the numbers are crunched, however, 
> two or three songs will cost $3 or $4, and the rest will, in a 
> special sense, be free--free because no one wants them.  (I 
> listened to Bob Dylan's "Freewheelin'" album today while 
> jogging:  15 for 15. What a batting average, what a bargain!)
>
> I don't believe on-demand purchasing will lower libraries' 
> costs. I believe that on-demand will better align the 
> expenditures to patrons' use.  If I may borrow (or abuse) a 
> phrase from economics, this is "hedonic" price adjustment, 
> where the quality improves even if the cost does not rise.
>
> We will inevitably be moving in this direction.  The operative 
> word is "inevitably."  Therefore in these very difficult times 
> for everyone, we should get started, when the motivation to try 
> just about anything is keener than in the fat times.  Seven 
> years from now we will regret not having taken action.
>
> Joe Esposito