The Average Voter Isn't a Dunce - She's Irrational

There are, of course, not many "average voters", strictly speaking. With that disclaimer (designed to somewhat allay the concerns of my readers who very much approximate the "average voter"), I present some of the findings of Bryan Caplan in his new book The Myth of the Rational Voter.  A teaser article is available on-line here:

The average voter (41% of Americans) believes that the largest part of the federal budget is spent on foreign aid. Apparently the average voter is including the cost of aiding countries by democratizing them by bombing them.

The average voter (40% of Americans) also believes that welfare is right up there with foreign aid at the top of the budget list.  This is also false.  Defense and social security together make up nearly half of the federal budget.  (And the average voter is not miscounting welfare as social security.)

The average voter (47% of Americans) thinks that having "too many immigrants" is a major reason why the economy isn't better off.  80% of economists would tell these Americans to get a clue, since "too many immigrants" is "not a reason at all".

What is Caplan's conclusion?  The right one, I think.  The average voter is an irrational nutjob when it comes to her political beliefs.  Caplan also provides a plausible explanation for this.  Having irrational political beliefs doesn't cost the average voter much.  If you have irrational beliefs about how your wife treats other men in your bedroom, you stand to lose quite a bit.  But holding the irrational belief that McCain is fit for public office doesn't cost people very much (unless, perhaps, all their friends are rational).  As Caplan puts it:

In a sense, then, there is a method to the average voter’s madness. Even when his views are completely wrong, he gets the psychological benefit of emotionally appealing political beliefs at a bargain price. No wonder he buys in bulk. -link

The claim here is that it is instrumentally rational for the average voter to be epistemically irrational.  That is, the average voter gets some non-epistemic value from being a epistemically irresponsible.  This non-epistemic value is more valuable to the average voter than is being politically and epistemically responsible, and so we have that the average voter is a loon (epistemically speaking).

Being an epistemic loon who votes is, of course, usually a terrible thing.  Epistemically irresponsible voters are, I presume, acting immorally every time they hit the polls.  (This will, for a variety of reasons, call into question Schwitzgebel's overcommon assumption - which I find absolutely absurd - that voting is a duty.  But I suspect he disagrees with several of the claims I've made.)  It does not follow immediately that an epistemically irrational voter acts immorally by voting, but it will follow if our irrational voter is in a strong enough position to recognize her own irrationality that she ought to recognize it.  And I think the average person is, in fact, in such a position.  So by ignoring her responsibility to be epistemically rational, while attempting to initiate force, I take it that she is acting immorally.  If this is right, it would be very nice if we could find an explanation for the average voter's utter cluelessness that does not appeal to her irrationality, as that explanation would be vastly more charitable.

The problem is, the irrationality hypothesis explains the evidence far better than any other hypothesis I'm familiar with.  Consider two other hypotheses: the miscalculation hypothesis and the ignorance hypothesis.  On the former, the average voter is, metaphorically, just bad at math.  On the latter, the average voter isn't irrational.  Rather, she's just too stupid to do any better.  But as Huemer notes here, neither of these hypotheses can explain (i) why political disagreements are so persistent, (ii) why people hold their political beliefs so strongly, (iii) why it is possible to predict, with a statisically significant degree of accuracy, a person's political beliefs if one knows their sex, race, income level, social status, etc..., and (iv) why political beliefs come in packages when there is no good epistemic or logical reason for the beliefs being so-packaged.  So much for those charitable interpretations.  How many others can you think of?