Math departments suffer, because they cannot afford buying the journals they used to buy. E.g., right now the famous Goettingen math department decided to cancel all(!!) its mathematical journals, in order to start over from scratch. But still, commercial publishers raise their journal prices by a rate which is unseen in most other economic areas.
Though it is possible to produce high ranked journals for as little as 15 US Cents/page or less (Annals of Mathematics, Documenta Mathematica) and to broadcast them for free on the Internet, there are journals charging 4 US Dollars and more per page!
If you want to find out about that, please look at the tables on http://www.mathematik.uni-bielefeld.de/~rehmann/BIB/AMS/Price_per_Page.html
Based on data collected by the AMS over many years, I have set up a collection of tables showing you the prices per volume and per page for 276 math journals. Here you can easily figure out which journal has the highest inflation rate over the years. In fact there are quite a few journals who have had an average inflation rate of 14 percent and more >>each year<< over a period of eight years or more! 40 journals charge more than one US Dollar per page!
Scientists are writing the content, scientists have developed and are maintaining the tools (like TeX) to format the content, and provide printer-ready work, do the peer review for free, give it for free to the publishers, and then buy it back from them for an incredible amount of money.
If you want to know how to do that: Publish in journals which are affordable - I have mentioned a few, and there are more. Tell your library to only maintain those.
If you are an editor, look at the SPARC publications: http://www.arl.org/sparc, for example at their brochure "Gaining Independence", a "Manual for Planning the Launch of a Nonprofit Electronic Publishing Venture", which tells you and your fellow editors what to do in order to make your journal independent!
November 2003, Ulf Rehmann (Bielefeld, Germany).