When people suddenly start changing their behavior in the same direction, economists presume that prices have changed. If we see that people are driving bigger cars, our knee-jerk guess is that the price of gas has fallen.
But there is a competing family of explanation: fads. If people like to copy other people, then we should expect behavior to change in the same direction even when prices stay the same. If some people start driving bigger cars on a whim, maybe other people will copy them, and away we go.
Alex Tabarrok recently linked to NameVoyager, a fascinating and user-friendly website about historical naming statistics. It is easy to verify that parents change their naming behavior in a correlated way: my parents' names were very popular in the '30's when they were born, and aren't popular any longer. And since we don't have to pay more for popular names, the explanation probably isn't changing prices. (Admittedly, there could be less obvious prices; Adolf dropped out of the top 1000 names in the 1930's).
So the explanation must be fads, right? I don't doubt that it's part of the story. But there are two counter-examples glaring me in the face. My wife and I picked the names Aidan and Tristan for our twin sons in 2002 in large part because we thought other kids wouldn't have the same ones. And what happened? The names almost instantly became more popular. Aidan lept from #311 in the '90's to #40 in 2004. Tristan went from #148 in the 90's to #116 in 2004. And I've talked to other parents with the same experience. They pick names they think are distinctive, but soon learn that they are unknowingly part of a big social wave.
I call this a "pseudo-fad." It looks like a fad on the surface. But if you study it more closely, you learn that at least a lot of people are acting independently, or actively trying not to copy others.
How is a pseudo-fad possible?
You might think that a common cultural shock hit us around the same time. Tristan was Brad's Pitt's character in Legends of the Fall, but that movie came out in 1994. It's a little hard to believe that this cultural shock incubated for eight years, then exploded. Then again, the name's frequency only moved from 600 per million to 850 per million, so the shock doesn't have to be large in absolute terms. Aidan is a bigger puzzle - it went from 200 per million to 2400 per million. And for the life of me I can't figure out why.