Measuring Home Preferences

Would you rather have a 4,000-square-foot house in a neighborhood of 6,000-square-foot McMansions, or a 3,000-square-foot home in a zone of 2,000-square-foot bungalows? I would choose the smaller home, mainly because I loathe McMansions and would not want to live in a neighborhood of them. Indeed, I bought an apartment in Manhattan, which is something that only someone who values small homes highly would ever do.

Many other people would think in similar ways when answering that question. But Robert Frank, as chanelled by Dan Gross, seems to think that the answer that people give has everything to do with wanting "to lord it over their neighbors".

Greg Mankiw says that the survey is wrong, in any case, and that people don't behave that way in reality. Alex Tabarrok actually runs the regressions and proves Mankiw right.

But it's not clear what the survey was actually measuring. If you wanted to look at the value of small homes in big-home neighborhoods as compared to big homes in small-home neighborhoods, you'd do that empirically, not by conducting a survey. So the survey was probably measuring what kind of home people say that they would prefer, rather than what kind of home they actually prefer. And those two things are rarely identical, or even particularly similar.