Dissecting Elsevier's "letter to the mathematics community"

Elsevier seems to have officially declared defeat in the face of the mathematical community's growing boycott to which I am party.  So, what now?  Let's look at Elsevier's surrender in more detail.  The boycott was based on three points, which I address below in reverse order.

3. Removal of support for the Research Works Act. This is a huge victory; since Elsevier's announcement, the bill has been withdrawn.  However, they have not changed their overall position on this front:

...we will continue to join with those many other nonprofit and commercial publishers and scholarly societies that oppose repeated efforts to extend mandates through legislation.

 2. Bundling of journals. Nothing on this front:

...we will seek to address concerns about the nature and composition of the large discounted agreements, through which most Universities now access journals – but addressing the base line pricing is a necessary first step.

1. High prices of Elsevier journals.

Our target is for all of our core mathematics titles to be priced at or below US$11 per article (equivalent to 50-60 cents per normal typeset page) by next year, placing us below most University presses, some societies and other commercial competitors.  Where journals are more expensive than this, we will lower our prices, as we already have in recent years for journals such as the Journal of Algebra and Topology and its Applications, among others.

The largest mathematics publishers (ranked by number of journals), and average price per article, according to data from journalprices.com*:

  1. Springer $33
  2. Elsevier $21
  3. Wiley $35
  4. Taylor Francis $55
  5. World Scientific $19
  6. Cambridge Univ. Press $20
  7. SIAM $8
  8. Oxford $22
  9. International Press $17
  10. American Mathematical Society $8

Clearly, if Elsevier were to lower the price per article of all of its mathematics journals to $11 or less, it would be one of the most economical sources of mathematics publications, trailing only our dear professional societies, SIAM and the AMS.  Unfortunately, I fear that the word core is a key detail here.  Elsevier publishes a whopping 69 mathematics journals; how many of them are considered core?  Assuming that "core" refers to the same 15 Elsevier journals whose archives are now open (see the list below), then the average Elsevier price per article over all mathematics journals will drop by about $2 (actually by $2.37), leaving the average at $19 -- not bad compared to its peers, but still outrageous for the minimal value added.  Of course, these numbers don't include journals like J. Comp. Phys. and CMAME that are (or were) very important to applied mathematicians.  I'm sure Elsevier is not aware that those would be of concern to mathematicians either.

Looking at the list above I wonder why we are going after Elsevier and not Springer or Wiley...

0. As a bonus, we got this:

...we have made the archives of 14 core mathematics journals open, from four years after publication, back to 1995...

 This means that the contents are available for free after 4 years (not after 1 year, which is the case for most other Elsevier open archives).  The letter fails to mention what the 14 "core mathematics" journals are, so I'll list them here (as found on this page):

  1. Advances in Mathematics
  2. Annals of Pure and Applied Logic
  3. Differential Geometry and its Applications
  4. European Journal of Combinatorics
  5. Finite Fields and their Applications
  6. Information and Computation
  7. Journal of Algebra
  8. Journal of Combinatorial Theory A
  9. Journal of Combinatorial Theory B
  10. Journal of Functional Analysis
  11. Journal of Number Theory
  12. Journal of Pure and Applied Algebra
  13. Linear Algebra and its Applications
  14. Stochastic Processes and their Applications
  15. Topology and its Applications

Interesting -- there are actually 15!  Perhaps one of them was already open?

*: Note that I've computed these averages by giving equal weight to each journal.  It would be more appropriate to weight them by number of articles, but I don't have easy access to that data.