Thursday, June 17, 2010

On the wide disparity in publisher cost-efficiency

Libraries and librarians are continuing to cope with the impact of the global financial crisis, and I understand that some of us are beginning to wonder why our libraries are facing deep budget cuts and staff furloughs while a few of the largest commercial publishers are boasting record profits.

Understandable, some are beginning to look for a better deal, and personally I think this is a very good thing. However, this might be a good time to highlight that there is a very wide discrepancy in the cost-effectiveness of different types of publishers. It it not fair, IMHO, to treat the mission-oriented publisher that has never charged more than they needed to survive, as if they were the same as the highly for-profit publishers. By all means, let's look for deals - but let's not forget that a 3% increase to a $100 subscription is only $3, while a 3% increase to a $1 million subscription is $30,000 - and the $3 increase might mean the difference to survival for the efficient publisher, while the for-profit would have to give up more than $30,000 to even begin to have something resembling a revenue cut of a much smaller order of magnitude than what is faced by many a library.

One extreme example from our own profession:

As of today, the subscription list price for the for-profit Library Management (Emerald) is EUR 11,819 (Ulrich's). That's $14,600 US (Bank of Canada currency conversion service, June 18, 2010). Compare this with the MAXIMUM institutional cost for ACRL's College and Research Libraries at $80 US for non-member institutions outside of the US and Canada. This is a difference of well over a hundred fold in subscription cost for these two journals, and I would argue that of the two, it is ACRL's College and Research Libraries that is the more prestigious. I do acknowledge that single subscription costs are of limited applicability in the world of bundled and largely consortial pricing. Also, this is an extreme example. Cost differentials in the range of 4 to 10-fold appear to be much more common (that is, the cost of one journal on a per-article basis can be a four to 10 times as much for another journal of similar or every higher quality. Imperfect though this example is, it is illustrative of the wide difference in costs between different publishers which does not necessarily correlate in a positive manner with quality.

The original analysis for this example is from my book, Scholarly Communication for Librarians, Chandos Publishing: Oxford, 2009. The cost differential has increased since the time the book was written; the ACRL cost is stable, while the Emerald cost has risen considerably.

[Disclosure: I am co-coordinator of the ARL and ACRL Scholarly Communication Institute webinar series, Building Strength through Collaboration, however this has nothing to do with ACRL publications or membership].

Heather Morrison, MLIS
The Imaginary Journal of Poetic Economics

This post was distributed to a number of listservs on Thursday, June 17, 2010

Update June 17, 2010 - this post by Bernd-Christoph Kaemper to Scholcomm and Liblicense-l is worth repeating in full:

Oh, well, Emerald again ...

Emerald (formerly MCB) has received its fair share of criticism in the past (including some from me, back in those days of the newsletter of serials pricing issues), but I've learned since then, that Emerald is an extreme example of pricing policies that appear to address the bold and adventurous among us, those that have no fear to apply all the techniques learned on the bazaar, rather than the timid, who ask for the list price and turn pale in horror.

To put it simple and bluntly, if you actually have a few Emerald titles in your collection that you pay at list price instead of a bundle, you or your predecessors in your job probably made a terrible mistake, or your institution has/had too much money to spend. Just cancel and start anew.

For some historical reason, that eludes my comprehension (although I'd love to hear explanations), single title prices with Emerald are fantasy prices. If you actually start talking to Emerald (and they really have friendly and competent staff), you'll be surprised to learn that it can be possible to get a package of titles for less than the list price of the most expensive title(s) in it, and that you actually might be able to get a rather decent offer for your campus or consortium. (Disclosure: my perspective and experience is that of a large technical university with comparatively small business and social sciences, but I have also talked to colleagues from other universities.)

Essentially, you buy Emerald as a full text database (various sizes and collections available) and have to judge its efficiency on that basis (i.e. on an aggregated cost per use base), not any individual titles a la carte which would be terribly expensive.

With respect to transparency and investigations of publisher cost efficiency and value for money on the basis of various possible metrics: I guess this is a prime example of why Ted Bergstrom, Paul Courant and Preston McAfee were right to ask libraries and consortia to disclose the actual terms of their bundle deals (under a public-records request) rather to rely on list prices, cf. e.g. the ARL press release Elsevier Motion to Block License Release Denied in Open-Records Decision,
and Ted Bergstroms Journal Pricing page,

I'll leave it to others who are in a better position to do this to possibly comment on the actual cost efficiency of Emerald compared to other publishers in economics on an aggregate basis, but my understanding is that this was also not the focus or intention of your actual message (to look at any particular publisher). I only wanted to give a little warning that individual title list prices have very little meaning here. (And please take also the bazaar analogy above with a grain of salt. The positive message here is: the publisher might be more responsive to your needs and constraints then you would have assumed.)

I'd also question your placative and simplistic contrasting of the (efficient) "mission-oriented publisher that has never charged more than they needed to survive" with the (inefficient) "highly for-profit publishers".

That's a poetic notion not one that adequately matches today's complex business realities for scholarly publishers, both for-profit and not-for-profit.

The question of cost-efficiency is a complex one, cf. for a starter the the JISC Report

Economic Implications of Alternative Scholarly Publishing Models: Exploring the costs and benefits (Jan 2009)

and the Feedback on it provided by STM and ALPSP, as discussed under

Best regards,
Bernd-Christoph Kaemper, Stuttgart University Library


This is very helpful information, thank you Bernd-Christoph. I should clarify that I do not mean to paint Emerald as a "bad guy" at all; in fact, Emerald has a very enlightened approach to green OA self-archiving policies. Nevertheless, I think that this fantasy price model - which I suspect is not unique to Emerald - bears further investigation. I plan to mull this over a little. On the other hand, ACRL and College and Research Libraries, producing top quality scholarly publishing at rock-bottom prices and making good efforts towards open access by supporting author self-archiving, and making many articles fully open access, is, in my mind, a clear-cut good guy.

This post is part of the Transitioning to Open Access series. From my perspective, a healthy open access future for scholarly communication should be built on sustainable business models.

Update July 4: for commentary on the fantasy pricing model, see Barbara Fister at Library Journal.