Think about the cafe at work. You have to wait too long for a coffee, someone forgets your regular order or you get served out of turn. And you swear that you'll never go there again. Umm, really? Does that happen?
The Freakonomics guys looked at the same thing in a different way with the NBA player strikes. Watching and attending sport provides utility to a consumer. When the strike goes on, the only threat the consumer can make is to reduce the future income of the league. But the league still provides the utility and making the change to some other sport involves big behaviour changes of the consumer.
As when Jim Carrey couldn't tell a lie but got bad service: "you know what I'm gonna do? I'm gonna shut up and take it in the bazooka because I can't do anything about it..."
Finally, there's a marketing and econometric model that represents it pretty well. I wrote this paper about five years ago and the data above shows a histogram for chocolate bars. The histogram shows a typical reverse J curve, many people don't buy a chocolate bar in a year, quite a few buy one, less buy two etc. This observed rate comes from two things: a consumer's long run purchasing rate and short term fluctuations around it. Professors Peter Fader and Bruce Hardie do a lot of this, and understand the workings way better than I do. Anyhow, there's fifty years of marketing science that indicates in general consumers don't change the volume of stuff they buy, and are pretty sluggish in changing their brand choices too. - Posted using BlogPress from my iPad