Is Christmas efficient? Jonathan Chait says it's not. Norm says he's missing the point. I'm with Norm.
The notion that Christmas is inefficient seems like common sense. I know my preferences better than anyone else. So I can spend money on myself more efficiently than anyone else can. So, if others buy stuff for me, it's likely to be on some things I don't value particularly highly. It would therefore be more efficient if we all spent money on ourselves, rather than each other.
In these papers (both pdf), Joel Waldfogel quantified this. He estimated that "gift-giving destroys between 10% and a third of the value of gifts."
End of, you might think.
Nope. In another paper (Jstor only) Sara Solnick and David Hemenway estimated that gift-giving actually creates enormous value. Recipients valued gifts at more than twice their price.
Why the difference? It lies in the question they asked.
Waldfogel asked: what is the "amount of cash such that you are indifferent between the gift and the cash, not counting the sentimental value of the gift?"
Solnick and Hemenway asked: "Aside from any sentimental value, if, without the giver ever knowing you could receive an amount of money instead of the gift, what is the minimum amount of money that would make you equally happy?"
There are two big differences here. Waldfogel uses the word "indifferent" whereas Solnick and Hemenway use "equally happy." The latter puts people in a better mood than the downbeat word "indifferent". This subliminally encourages people to value things more highly.
Also, Waldfogel asks people to disregard the sentimental value of the gift at the end of the question, whereas Solnick and Hemenway do so at the start. Waldfogel's question therefore emphasizes the disregard more, giving people a greater focus upon the mere material features of the gift.
I take two lessons from this. First, as Bradley Tuffle and Orit Tykocinski say here (pdf), "subtle changes in wording can drastically affect subject response." Surveys are therefore a tricky way of elicting reliable information. Survey design is a great skill.
Second, it's sometimes harder to prove that common sense is right than you might think. This might be one reason why so many people prefer armchair theorizing to hard empirical work.
Above all, though, we should remember that there's something odd about ignoring the sentimental value of gifts because, as Norm says, this is precisely the point of them.
There's more to life than instrumental rationality. And if you forget this, you misunderstand the nature of rationality.