Mankiw's textbook is a giant leap backwards in the teaching of economics to undergraduates. Naturally, it is always commendable to attempt to simplify complicated concepts to students, but there are limits - particularly in a principles textbook, perhaps the only economics that many people will ever encounter. In this case, it can be considered downright dangerous.
Mankiw "dumbs down" economics, underestimating the average intelligence of a university student. The clearly anti-analytical tone of the book can be counterproductive at times. Several comments from my students noted that Mankiw's style makes it seem that, say, a particular argument is "obvious" when Mankiw explains it, but they could not produce it by themselves since the argument does not seem to use any of the tools which they are busily acquiring. In other words, what is "obvious" to Mankiw and to someone with more training in economics, is rarely "obvious" to the student. The student with analytical bent who has made the necessary investment into learning the tools will find them useless in a lot of the problems set out in the book.
The irritating "scientistic" tone of the book makes it seem as if what Mankiw says is "the truth" and "accepted" by all economists. At least Barro's principles text was considerably more honest in its partisanship and announced loudly that there were other views. Mankiw does not do this. What makes this "scientistic" tone even more irritating is that it is used with particular verve in areas where current economic debate is most controversial - thus giving a misleading view of even the "mainstream" of economics. Furthermore, it makes a lot of economics boringly "deadpan" rather than a living and growing discipline.
Most remarkable, given Mankiw's supposed specialty, is the incredibly compressed and poorly-written last few chapters on macroeconomics. He has singlehandedly decided to obliterate 70 years of research in macroeconomic analysis and leap back to the empirical platitudes of the 1910s and 1920s. It is one thing to be anti-Keynesian, it is quite another to pretend that the Keynesian Revolution never happened. Not even Barro did that. What is most dangerous is that Mankiw is now teaching thousands of students that unemployment is caused by labor market frictions and high wages, and never a word on effective aggregate demand (which, until Mankiw's textbook, was THE mainstay of macroeconomics).
On style: Mankiw writes in a nice conversational tone, albeit a bit too elementary, and always with very carefully chosen words. His examples are not always well chosen and even disturbing at times (e.g. the initial discussion of the impact of seat-belt laws).
I used this book to teach elements of macroeconomics this year. It was a terrible choice and I shall never do so again. Perhaps elements of microeconomics may be better served. Mankiw's textbook is not worth acquiring, assigning or reading. Barro or Colander are much, much, much better - and far more interesting.
