Perhaps the greatest truth about human nature that you do not find in the typical economics textbook is that people are sheep. Most human beings don't like to be different from the others around them; they want to fit in - to look, sound, and act "normal." (Perhaps the best discussion of this is in Gaetano Mosca's The Ruling Class; unfortunately, it is apparently not yet online).
The urge to be normal is most obvious in K-12 education, when conforming to changing fashions becomes a full-time job. But adults are basically the same, just a little more subtle. How much would you have to pay typical adults to wear shorts at a formal gathering of people they are sure to never meet again? How much would you have to pay the typical man to wear a dress?!
But if people are sheep, doesn't it undercut a great deal of economics? Economists are always arguing that you can change behavior simply by changing prices. But if people are desperate not to be square pegs, they might lack the courage to switch to currently unusual behavior just because the pay is better.
In fact, I'll warrant that if you pulled people at random out of a crowd and offered them decent money to do something weird, most wouldn't do it. That's precisely what an interesting paper in Public Choice by Werner Guth and Hannelore Weck-Hannemann found. They gave randomly selected students a chance to sell their vote for a fair bit of money. And even though many of those students didn't vote, very few were willing to publicly take the cash.
It is not too difficult, however, to reconstruct economics on a "sheepish" foundation. All we need to realize is that there are degrees of sheepishness. In any society, 5 or 10% care primarily about money, not what other people think about them. (Or perhaps they reason that once they're rich enough, people will think what they want them to think!) When prices change, the money-focused people change their behavior. That makes changing behavior less weird, which makes slightly more sheepish people willing to change their behavior. And that reduces the stigma further, and so on.
Unless the initial stigma is overwhelming, then, it is easy to get big swings in behavior even though most people are suckers for peer pressure.
Thomas Nechyba's model of welfare and illegitimacy is a nice example. Who would have thought that paying unwed moms for out-of-wedlock births would have led to a lot more of them? Was AFDC really such an attractive deal, considering all the social disapproval unwed moms still faced? The weakness in this argument is that some women were swayed by the money, which reduced the stigma, leading more potential moms to change their behavior... and off we go.
Economics should recognize that people are sheep because it's true. But I believe it would also improve economic pedagogy. When economists explain the effects of rent control, for example, it can sound a little paranoid. First, rent control... and before you know it, landlords are setting fire to their own buildings! A more plausible story is that rent control leads a few landlords to unscrupulous behavior, which makes it seem more normal...