Last night I took part in the now traditional Halloween economic experiment at Dean Karlan’s house. We were handing out candy to trick-or-treaters, but with a twist. First, we asked the kids if they were willing to donate the candy we gave them, either to the Connecticut Food Bank or to Freedom From Hunger. Second, we had a team of Yale students and IPA staff sitting alongside frantically tapping the data – on kids’ ages, costumes, and candy decisions - into laptops. Are children more likely to donate to a charity in the US or overseas? Does creativity in costume choice correlate with charitable contribution? Much remains to be seen. I spotted a definite age effect. The young-uns aren’t giving up their sweets for anyone. But kudos to the kid who emptied her entire evening’s haul into our charity bucket.

You can download the papers from previous year’s experiments on the "Halloween" section of Dean’s website – with findings that correlated behavior in the experimental game with non-laboratory behavior (costume choice), and that in the run-up to the 2008 Presidential Election, kids who picked the Obama candy table were more trusting than those who picked the McCain candy table.