This book represents a considerable revision and expansion of Public Choice II (1989). As in the previous editions, all of the major topics of public choice are covered. These include: why the state exists, voting rules, federalism, the theory of clubs, two-party and multiparty electoral systems, rent seeking, bureaucracy, interest groups, dictatorship, the size of government, voter participation, and political business cycles. Normative issues in public choice are also examined. The book is suitable for upper level courses in economics dealing with politics, and political science courses emphasizing rational actor models.
Mueller's Public Choice is a monstrous, 800 page "overview" of the field of Public Choice. Credit where credit is due, unlike the previous book on the subject I've read, this actually does do a good job of explaining what the field is about and covering the main ideas.
But who really needs an 800 page overview? For students, the book is of little use as it doesn't actually teach anything. For scholars, you'd be interested in a couple of chapters at most, and the detail in those chapters is sacrificed in favour of the others that you won't read. A potential use case is for a PhD student trying to find a topic to study, but in that case the fact that this book is already twenty years old will probably be critical.
This was very hard to read. Not because of the size (although there are the 700-odd pages to go through), not because of the mathematics and not because of the style of the author (which could be better). It was mostly because the implications of what you read need time to be thought out.
Two simple examples for that are Condorcet's and Arrow's theorems.
The book is mostly an overview of all the literature on the topic of public choice - e.g. politics and economics of making decisions with voting. It contains some very good chapters on voting systems, on structures, on basic game theory.
The bad part is that the book contains too much theories which somewhat don't deserve the name and concern spherical horses in vacuum. The rational, maximizing economical subject that's used too much is practically not found in the real world, and although it helps for some models, you should always keep in mind that the real world is extremely complex, and a lot of the models don't do it any justice.