The basic content of intermediate micro - choice theory, theory of the firm, market equilibrium - has not changed for decades. The latest Intermediate micro textbooks cover substantially the same material - indifference curves, budget constraints, cost curves- as do 20 year old texts.
Sure, there are some differences across editions. Take, for example Eaton, Eaton and Allen's intermediate microeconomics textbook. The application “Why Everyone Loves Britney Spears” in the 6th edition was changed to “Why Everyone Loves Christina Aguilera” in the 7th edition (the page number, p. 272, is unchanged).
Or consider Browning and Zupan's Microeconomics. Previous editions focused on the inefficiency of food stamps and the virtues of school vouchers. The latest one (11th edition) has a new discussion of ObamaCare: "We have no expertise in constitutional law, but it is easy to use our consumer choice model to show that a mandate combined with the subsidy can make some recipients worse off than they would be with no subsidy at all."
Publishers produce new editions of undergraduate textbooks every three years for one reason: to kill the second hand market.
The Carleton bookstore is charging $134.50 for a used version of Browning and Zupan (11th edition), $179.25 for a new one. Prices on Amazon are only slightly lower. Prices at this level are only sustainable if there is a relatively small number of used texts available - if there were 10 or 20 years worth of used textbooks on the market, the increased supply would drive down the price of both old and new texts.
Now there are 10 year old copies of Browning's Microeconomics available - a 2001 version sells for $1 plus shipping and handling through abebooks.com. The reason that there is such a huge price disparity between old and new editions - although the books have substantively similar content - is network externalities. Textbooks serve a coordination function; they allow students and professors to be "on the same page", both metaphorically and literally.
Professors generallly assign readings by chapter, for example, "Week 2: chapter 4, consumer theory." But chapter 3 in the current edition might be chapter 2 or chapter 4 in the old edition. Professors often require students to complete end-of-chapter problems, and problem numbering or wording may differ across editions. When a professor says "Study Figure 3.4 carefully, I'll be asking a question about it in the final exam", the student with an older edition of the text, where Figure 3.4 is labelled 2.4, may end up studying the wrong thing.
It's just like software upgrades - an old version of Microsoft Word might do everything the user needs, but if Microsoft introduces a new .docx file format, old versions are rendered useless, because they are unable to read files generated by other users.
The whole process disgusts me. New editions destroy value, by making old editions worth less. They hurt students, who face high prices for texts. Any pedagogical benefit from having a slightly more updated text is more than offset by the increase in the number of students who decide to save money by reading Wikipedia instead of a textbook. Ultimately, it is a collosal waste of resources - all the effort that goes into producing a new edition that differs little from the old one; all of the perfectly good older textbooks that end in the recycling bin.
But what is the alternative? One possibility is banning new editions - or limiting publishers to one every six years.
Alternatively, professors could force publishers to compete on price by giving their students a choice of texts (e.g. Readings - pp. x-y, Varian OR pp. y-z, Eaton OR pp. a-b, Browning). The obstacle this solution faces is that professors like to rely on publishers' add-ons, for example, end-of-chapter questions, and it's impossible to require students to complete end-of-chapter questions if every student has a different textbook. Coordinating with students when every one has a different textbook is difficult.
There are a couple of micro textbooks available free (legally!) on-line, e.g. David Friedman's price theory text, or Rittenberg and Tregarthen's Principles of Micro. Neither of these are ideal for the type of intermediate micro course I teach, but the Rittenberg and Tregarthen model - where the text is provided free, and revenue is generated by the add-ons - seems to me to have considerable promise.