How Irrational Are Voters?


Some interesting press about a very interesting new book by Bryan Caplan, an economics professor at George Mason University: The Myth of the Rational Voter: Why Democracies Choose Bad Policies.

The New York Times Magazine gives it a once-over, summing up its message thusly:

Bryan Caplan, an economist at George Mason University, has attracted notice for raising a pointed question: Do voters have any idea what they are doing?….Caplan argues that "voters are worse than ignorant; they are, in a word, irrational — and vote accordingly." Caplan's complaint is not that special-interest groups might subvert the will of the people, or that government might ignore the will of the people. He objects to the will of the people itself.

In defending democracy, theorists of public choice sometimes invoke what they call "the miracle of aggregation." It might seem obvious that few voters fully understand the intricacies of, say, single-payer universal health care. (I certainly don't.) But imagine, Caplan writes, that just 1 percent of voters are fully informed and the other 99 percent are so ignorant that they vote at random. In a campaign between two candidates, one of whom has an excellent health care plan and the other a horrible plan, the candidates evenly split the ignorant voters' ballots. Since all the well-informed voters opt for the candidate with the good health care plan, she wins. Thus, even in a democracy composed almost exclusively of the ignorant, we achieve first-rate health care.

The hitch, as Caplan points out, is that this miracle of aggregation works only if the errors are random. When that's the case, the thousands of ill-informed votes in favor of the bad health plan are canceled out by thousands of equally ignorant votes in favor of the good plan. But Caplan argues that in the real world, voters make systematic mistakes about economic policy — and probably other policy issues too.

The Times' author Gary J. Bass misses one of the book's most interesting ideas: the principle of "rational irrationality." This goes beyond the standard public choice concept of "rational ignorance" (where voters aren't educated on policy because it doesn't pay off personally for them to be so) to point out that, since people get some pleasure out of certain irrational beliefs, and in a voting process in which we have almost zero chance of affecting the outcome it doesn't particularly cost us anything to indulge them, people have economically sound reasons to not bother being rational when they vote. Indeed, they don't bother doing so, he maintains (with scads of evidence) in his book. I first referred to this concept of Caplan's in the pages of reason in this May 2003 review essay about public opinion research and democracy.

A review of Caplan's book in the lefty mag In These Times is in many ways a more thorough explanation of what Caplan has to say (author Christopher Hayes does understand and explain the centrality of "rational irrationality" to Caplan's thesis) and, despite ideological differences, gives some kudos:

It's tempting to dismiss Caplan's thesis out of hand, because it's so self-consciously "provocative" and because he's translating an old discredited anti-democratic argument into the jargon of econocentric elite-speak. But if you support democracy, you must confront the fact that voters can often be stunningly under-informed and that majoritarianism run amok can lead to persecution, hatred and injustice. Reading Caplan's book, then, is both bracing and necessary because it forces the reader to stare into the abyss—an abyss the author seems only too happy to jump into.

Caplan blogs at Econlog. He contributed to a reason interview roundtable on the Federal Reserve in our November 2006 issue.