The Swing Voter's Curse: Comment

Abstract

In an important paper, Timothy J. Feddersen and Wolfgang Pesendorfer (1996) investigate the “swing voter’s curse.” In a model of elections in which voters have common preferences and private information, they show that, when indifferent, less informed voters are better off abstaining than voting for either alternative, even without a cost of voting—the “swing voter’s curse.” They go on to show that even though significant abstention occurs in large electorates, the outcome of the election is almost always the same as with perfect information. Unfortunately, the proof of the proposition that establishes the “swing voter’s curse” phenomenon contains an error. In this comment we identify the error and give a correct proof of the proposition.