More on voting.

• Author: Bohanon, Cecil E.
• Position: Reply

Public-choice analysts have argued for a long time that the probability that an individual's vote will matter in an election with many voters is near zero. Nevertheless, in the brouhaha that surrounded the U.S. election of 2000, media pundits repeatedly claimed that the election demonstrated clearly that "your vote matters." Our article "Now More Than Ever, Your Vote Doesn't Matter" in the spring 2002 issue of The Independent Review (591-95) argued that, on the contrary, the 2000 election actually reenforces this public-choice insight. In this issue, however, Jac Heckelman offers several constructive criticisms of our analysis.

First, Heckelman correctly points out that we erred in saying that "only Florida's 25 electoral votes and its six million voters were decisive ... the votes of the other 99 million were not decisive" (Bohanon and Van Cott 2002, 591, emphasis in original). As he notes, all states that voted for Bush were equally decisive in the election because of the slenderness of the electoral margin. The analogy might be a basketball game won by one point. Some might label the winning team's last-made shot as decisive, but any score the winning team made during the game was equally decisive. Florida's electoral vote was decisive only in the sense that it was contested--not, as we incorrectly claimed in our article, that Florida's electoral votes mattered more than other electoral votes for Bush.

We should remember, however, that the electoral college's actions are largely ceremonial. Scholarly interest in the incentives facing the college's electors is minimal. Each state's electors typically vote robotically for whoever won the presidential vote in their state, even though the U.S. Constitution does not bind electors' votes (nor do most states bind their electors). Therefore, any investigation of the "decisiveness" of electoral college voting ultimately must turn on the incentives of individual voters in each state.

Heckelman's second comment deals with these individual-voter incentives. Our article suggested a "recount-triggering" model as a superior alternative to the "tie-breaking" model in the public-choice literature. We claimed that "the probability of [a recount] occurring is the same as the probability of breaking a tie." Heckelman correctly demonstrates that an individual's vote matters less in a recount-triggering model than in the typical tie-breaking model.

A simple numerical extension of Heckelman's second point may...