Scholars from Mancur Olson to Anthony Downs have claimed that the act of voting is irrational. Their arguments are based on a weighting of the potential for casting the decisive ballot versus the costs of voting. This notion is central to rational choice theory’s description of voting behavior.
Where rational choice goes wrong is in positing that the benefit of voting (or expected utility) is captured only in casting the decisive ballot. People vote for a variety of reasons that are subject to change over time. In other words, a voter does not place a fixed uniform value on voting. Below is a discussion of some of these reasons and also a suggested measure for determining voters’ collective expected utility.
It would be foolish to posit that any value other than influencing an election as a primary motivator for voting. If this was the sole factor of a person’s decision whether or not to vote, we could accept the rational choice arguments, and scratch our heads and wonder just why so many people engage in behavior that is so obviously irrational. But there is, obviously, more to this story.
People vote not only to influence an election, but to exercise a fundamental right. They place some inherent value in the exercise of this right. In fact, some people at some times place this value above that of influencing an election. For example, people who vote in an election in order to cast a vote for a candidate with no legitimate shot at victory. The only way to explain this behavior is to assume (a) people are irrational dupes or (b) the inherent value of exercising this right is greater than whatever costs attach to the act of voting (direct and opportunity).
Another value that drives voting behavior is symbolic. A protest vote would be one example. If candidate X was extremely likely to win, what would motivate someone to take the time and effort to vote for that candidate? It is the symbolism of the protest vote- of saying that, “I’ve had enough of candidate X or their party.”
People also engage in voting for social reasons. These may include peer pressure, signaling, desire to share in psychic benefits derived from supporting a winner, etc.
It is also important to note that these reasons are not mutually exclusive. In fact, I would argue that very little voting behavior is explained by one distinct benefit. And, these expected utilities are prone to change over time, not only in terms of their interaction effects, but the discrete value of a particular factor.
As an example, in America we see the pattern of voting becoming more common with age. Some of this may be attributable to diminished opportunity costs, especially with regards to retired seniors. But then what explains increased voting of non-seniors versus younger groups? Could it be that the intrinsic value of voting increases with maturity?
Finally, the rational choice theorists posit a very crude cost benefit analysis of expected utility (as a product of potential for casting efficacious vote) versus the costs associated with voting. But I believe there are at least two problems here. First, as mentioned already is that expected utility is dependent on things other than an efficacious vote (or perhaps we ought to enlarge the definition of efficacious to include these others). Also, we can measure the utility voters receive in ways other than examining the efficacious vote.
One method would simply be to calculate the costs of voting, and then assume that the utility is greater (else we’re a society of irrational dupes). Or, a more robust method of contingent valuation. Now, I do not believe that a willingness to pay construct would work (at least here in the US). Voting is a fundamental right and asking people how much they would be willing to pay to exercise that right would skew towards an undervaluation. However, a willingness to accept procedure might get us to a more reliable calculation. Or, we could use both methods and have an upper and lower bound for our estimate.
Regardless, the notion that the act of voting is irrational is antiquated and easily falsifiable.