1. New editions are not churned out every six months.
2. In most cases you can get away with not having the latest edition for a class . Schools don't change the syllabus just because a new edition of a text book has just rolled out.
And the section numbers, case studies and questions get shuffled around too, so the previous edition is not sufficient.
The market is inefficient because the folks who choose which books to buy - the people who set the course - are not the people who have to pony up the costs. So it's in the interest of the publishers to maximize their gouging of students, and then spending a good fraction of the surplus on "incentives" for course setters to keep them on board and recommending the latest edition. The publisher who gouges the most has the most money to spend capturing these interests, creating a feedback loop, so for any given course vertical, there's a tendency to a monopoly.
I believe this eliminated the majority of calculus/linear algebra/diff-eq texts from introductory courses, so it has become easier to find used copies.
Don't mean to be snarky but I get the impression that the relationship between US textbook publishers and academia is really the worst of all worlds, and seems like a case of the free market gone wrong, with few colleges working to offer a competitive (ie cheaper) option to students who don't want to pay ridiculous prices for books.
Mind you, it's great for autodidacts like me who can pick up slightly older textbooks for less than the cost of printing. Now if only Spivak would do a new edition of his calculus book...