Richard Thaler was one of the inadvertent founders of the field of behavioral economics. I say "inadvertent" because it wasn't clear to Thaler or his colleagues exactly what he was doing when he started looking into what he called "The List": discrepancies he'd noted between the predictions of economic theory and the choices people actually make. Those discrepancies profoundly violated the principles of traditional economics—and, crucially, did so in ways that weren't attributable to random error.
Here's an example: Thaler gave half of the students in one of his Cornell classes a university coffee mug. The students with a mug were then asked to write down the minimum price at which they would sell it, and the students without a mug the maximum price at which they would buy one; sellers were then matched with buyers. According to classical economic theory, if the mugs were distributed randomly about half the students with a mug should value them less than students without one, and so roughly half the mugs should change hands.
You may be thinking, "Who cares about a coffee mug?" But that's precisely the point: no one cares much, unless they already possess one. Simply owning a mug, even when they didn't pay for it, even when they had owned it for only a few minutes, made students value it more. This is a violation of a key tenet of rational choice economics, in which preferences (in this case, for owning or not owning a mug) are assumed to be stable.
The coffee mug experiment illustrated some key principles of the then-emerging field of behavioral economics—that is, economics that tries to take our observable behavior into account, instead of treating us like the super-rational, utility-maximizing, optimal-choice-making agents of traditional economic theory. (Thaler calls these theoretical beings "Econs.") The first principle is loss aversion: losses are more painful to us than gains are pleasurable—in fact, Thaler's colleagues Daniel Kahneman and Amos Tversky found that losses give us at least twice as much pain than gains of the same amount give us pleasure. (I've also written a post on Kahneman's fascinating book Thinking, Fast and Slow (Farrar, Straus & Giroux, 2011).)
The second principle is status quo bias: we are resistant to change. We tend to stick with what's familiar or what requires the least effort, even when we recognize that there may be benefits (in the mug experiment, cash) if we're willing to change. Together, these behaviors result in what Thaler calls "the endowment effect": our tendency to value things we own more highly than things we don't—and more highly than things, including cash, that we could exchange them for.
The endowment effect depends on our ignoring what economists call "opportunity costs." As an example, imagine that you're a sports fan who has just been given a free pair of tickets to a key playoff game. You might be able to sell the tickets for $1500, but many fans would choose to go to the game—after all, the tickets were free, weren't they? To an economist, though, there's no difference between using free tickets worth $1500 to go to the game and paying $1500 out of pocket to go to the game, since you could have had that money if you had sold the tickets.
That almost no one but an economist thinks this way, though, did not seem to bother economists for about two centuries. Instead, economic models that assumed hyper-rational behavior came to determine much policy and law. But it turns out that many of our actual behaviors are, as an economist would say, sub-optimal. The status quo bias, for example, means that when we are required to opt in to beneficial programs, such as employer-matched 401(k) accounts, too often we don't.
Thaler co-wrote a book with legal scholar Cass Sunstein entitled Nudge: Improving Decisions about Health, Wealth, and Happiness (Yale University Press, 2008), which used the insights of behavioral economics to argue that instead of requiring people to opt in to beneficial programs, we should use the status quo bias for good and instead require them to opt out. Changing the default option can radically change participation rates.
But, of course, there's no reason why the principles of behavioral economics can't be used for other purposes as well, and they are: by political groups, advertisers, and companies trying to shift your choices in ways beneficial to them. Misbehaving is an entertaining way to increase your awareness of how, and how easily, we can be manipulated. And with that knowledge, perhaps, we can try to make our choices—political, social, and economic—more conscious ones.
Update 5 Oct 2015: Thaler has created a Misbehaving website which includes outtakes (passages cut from the final manuscript), resources for learning more, "characters" (colleagues mentioned in the book), and a blog discussing real-world applications of behavioral economics.