Phishing for Phools has one big important idea. The idea is that deception is fundamental to market economies - that it's not just the result of certain models or certain situations, but that it's always present to some degree. The reason is that companies are always looking for new ways to trick people out of their money - to find new information asymmetries, bounded rationality, suboptimal behavior patterns, and legal loopholes to exploit. Given that companies are always looking for these, their will always be some degree of trickery and mistakes in any market. This is just an obvious result of costly monitoring (though I'm not sure Akerlof & Shiller actually mention that explicitly). They call this a "phishing equilibrium".
I wish the book had spent a lot more time on this idea. Obviously not all of our economic interactions consist of trickery and mistakes (right?). What affects how much or little "phishing" an equilibrium entails? How can we detect which of the different kinds of phishing ("informational phishing", "behavioral phishing", etc.) is going on? What are the principles of market design that minimize phishing? How can one detect, empirically, how much phishing is present? How can individuals learn to better avoid phishing?
Sadly, the book mostly eschews these questions and spends almost all of its pages telling anecdotes about situations that the authors think involved lots of phishing. The 2008 financial crisis, the Vioxx recall, political lobbying, advertising, and so forth. This is sort of interesting, but a lot of these are stories we've heard before. And they usually don't cite much research verifying that phishing was indeed central to these screw-ups.
That's a shame. That research often exists. I've reported on it. But instead, Akerlof & Shiller sort of assume that the stories speak for themselves - that any reasonable person listening to these tales would instantly realize that phishing was the only explanation in each case. I wish the authors had seen fit to include a little more proof and a little less rehashing of well-known events.
Another problem I had with the book was the tone. It's full of the stilted phrasing of academic econ papers - "In the following chapter, we will show", and so forth. I think that's just bad practice for a pop book - the general public is probably turned off by that lingo, and economist readers don't really need it. Phishing for Phools was not nearly as bad in this regard as The Assumptions Economists Make (shudder). But if you're an economist writing a popular book, just go ahead and ditch the "we will show"s and the "it has been demonstrated that"s.
And what on Earth told Akerlof and Shiller that calling people "phools" was a good idea? No one's phooled by that "ph", you know (especially not in the audio version). You're calling people fools. You're saying that they're too irrational and/or dumb to make it in a market economy. Even if that's true, isn't there some nicer way to say that? I mean, presumably Akerlof and Shiller would like people to support government policies that curb "phishing". But if people think they're admitting stupidity by calling for those policies, I bet they'll often oppose the policies just to save face.
Anyway, to sum up, I think Phishing for Phools is an interesting book that injects one very interesting, very important idea into the economic discussion. But I think that the choices of topic and tone lessen the effectiveness of the message. We need both more academic econ papers and catchier pop books about this topic.