Each October 31st in the US, little rapscallions and rug rats dressed as witches, demons, and other spooky nasties go door to door to screech "trick or treat" in hopes of getting candy. Without fail, said rapscallions and rug rats receive some candy they don't like (like the year-old Lemonheads and other tricks treats I tossed out last night). This phenomenon associated with gift-giving is hardly unique to Halloween.
Joel Waldfogel examined the deadweight loss of Christmas in his AER article from 1993: the social loss that arises when the amount by which spending on a gift exceeds the recipient's value. You see, I may have been willing to pay as much as $5 for that, well, thing given to me by my late aunt Bertha, but she spent $10 on it. The $10 represents the opportunity cost of the resources used in making that thing, so Bertha's giving me the gift, in effect, made society poorer.
Part of the difference is explained by sentimental value. That thing was worth only $5 to me, but the fact that my dearly departed Aunt Bertha gave it to me is worth $2 to me and her giving it to me was worth another $1 to her. So the total value is $8, meaning the loss to society is only $2.
But there is no such value with Halloween candy because it is largely given to kids by total strangers. Suppose I go to old man Withers' house (he runs the haunted amusement park) and he gives me Baby Ruth candy bars. Since I don't like coconut, I wouldn't be willing to spend anything to get one. And because I don't give a rat's patootie about Old Man Withers and he doesn't give a rat's patootie about me, there is no sentimental value. The difference is pure deadweight loss.
Yesterday I asked a parent what he and his three kids were doing for Halloween. He said that his kids don't want to go trick or treating. Instead, they want to go to Old Country Buffet. My first thought was, WTF? OCB? OMG! Y R U going 2 OCB? But at least there won't be any deadweight loss.
Leave it to an economist to point out the bad things in life!