Bryan Caplan's The Myth of the Rational Voter

In The Myth of the Rational Voter, Bryan Caplan tries to answer the question: Why do public economic policies go against the public's interest? For example, why is there so little support for true international free trade when economists have shown it to improve wealth? Why is there such support for farm subsidies when they are expensive and mostly go to wealthy corporate farms? Why are gasoline price controls a viable subject of debate when most living Americans lived through such a disaster in the 1970's?

Caplan's answer is simple and controversial: the common voters are simply irrational. There is strong evidence that voters' beliefs on economic policy significantly diverge from those of professional economists on factual matters. However, in the face of this evidence, there has been a long-held bias among scholars who study voting behavior and the democratic process, sometimes implicit and sometimes explicit, in favor of assuming rationality in the voting booth. And the bias is strong enough to preclude coming to such a conclusion (if, that is, you wish to be published).

It was mainly economists who first started to criticize democratic processes with the emergence of public choice economics after the mid 1900's. Pro-democrats had held that democracy does the voters' will, and that led to policy in the public's interest. Public choice came along and punctured this idea - voters would rationally choose to stay ignorant of what makes good policy because their chance to sway an election was essentially zero. There exists no good incentive to understand economics and policy, leading to capture by special interests; thus, democracy does not do the voters' will and often leads to policy that goes against the public's interest.

At a first glance, the public choice assumption of rationality made some sense - this, of course, was a school of thought developed mostly by economists, and the assumption of rationality had served them well in modeling and predicting human behavior in the consumer marketplace.

However, as Caplan shows, the assumption when applied to the political marketplace fails emperical testing. Frankly, the voting public has wacky ideas about what makes good economic policy. In other words, they are simply irrational. Furthermore, Caplan shows that this irrationality should be predicted by economic theory, dubbing his unique theory "rational irrationality". As the price of irrationality falls, consumers purchase more of it. And in the voting booth, that price is essentially zero. Therefore, the presence of so many irrational voters will lead to democratic failure in the form of unwise economic policy.

So the pro-democrats are correct that democracy does the public will. But public choice is correct that democracy will go against the public's interest. But it is precisely because of the democratic response to the public will, not in spite of it. In the end, H.L. Mencken was right when he said, "Democracy is the theory that the common people know what they want and deserve to get it good and hard."

From this basic groundwork, Caplan details the evidence that illustrates the difference of opinion between the public and academic economists, he lists the specific reasons that cause the irrationality, and takes a few stabs at remedies for this problem. The work has its weak points, but overall it is especially entertaining in the author's playful style and original point of view. It was positively reviewed in The New York Times, The Economist and many from all over the internet have engaged in the debate over the books conclusions. You can listen to Caplan discuss his work in an interview at EconTalk.

Bryan Caplan is a decided favorite of the many bloggers at the Distrbuted Republic, and this author is no exception. He does interesting and unique research, he has contributed to our annual May Day remembrance and has always been gracious in engaging in debate with us. It was my distinct pleasure to have an email chat with Caplan recently:

Trent McBride: How did you first get interested in economics in general, and public choice specifically?

Bryan Caplan: I know it's a cliche, but it began with Ayn Rand. Her non-fiction directed me to the Austrian School – I went to the Mises Institute Summer Seminar at Stanford in 1989 right before I started my freshman year at UC Berkeley.

I don't remember when I became interested in public choice – it was just one of a dozen fascinating branches of economics. But reading Donald Wittman's The Myth of Democratic Failure set me on my current course – he was asking the kinds of questions I wanted to answer.

TM: In the book you discuss the bias against scholars who do not assume voter rationality. Have you personally been denied publication because of assumptions about rationality?

BC: Yes, I've had a couple of referee reports say so explicitly. I suspect it was a factor in many other rejections, though it's hard to be sure. But I'm not one to complain. In the end, I've managed to publish even my "weirdest" research. In fact, my weird research has been markedly more successful than the "plain-vanilla" papers that came out of my dissertation.

TM: Traditionally, libertarians have aimed their collective ire at politicians (along with the media and the academy). Your book seems to claim that the wrong suspects have been fingered - to an extent, all three are products of the public's irrational beliefs. Do libertarians need to re-examine many of the common arguments in our arsenal?

BC: Factually – definitely. If voters changed, politicians and the media would rapidly adjust. (Academia would hold out longer; tenure gives a lot of slack!)

Rhetorically – probably. If people favor farm subsidies, it doesn't make a lot of sense to tell them that their politicians must be "corrupt" because they favor farm subsidies. You would probably gain more ground by arguing against farm subsidies on their merits instead of criticizing politicians motives. But I could be wrong about this - I'm the first to admit that I'm no master of political rhetoric.

TM: In an interview, Tyler Cowen stated that he would like to "rewrite public choice economics, putting the particularities of culture at the forefront and not treating them as a second-order effect." If you were to write this book, by having made rational irrationality the focal point of your theory of democratic policy outcomes, would you have to explain differences in policies across countries (or, at least, across democracies) by citing differences in rationality (or irrationality)? If yes, how would you do that; if no, what else in cultures could explain differences in democratic outcomes?

BC: As far as I know, no one has the data to test this claim, but I am convinced that democracies with more rational electorates have better policies, and vice versa. Despite their anti-market tendencies, Americans are a lot more pro-market than, say, the French. Indeed, Americans today are a lot more pro-market than Americans in the '30's.

TM: Could the theories in your book be applied to other forms of government? Is there evidence in non-democratic countries that policies still match the views of the public?

BC: Good question – I actually address it in a forthcoming paper in the Review of Austrian Economics. Short answer: Some dictatorships get very far away from citizens' preferences – almost no Russians wanted Stalin to seize their land and force them into collective farms. But most dictatorships avoid unpopular policies. Indeed, we often see that bad policies imposed by dictators turn out to be politically viable under democracy. Eastern European democracies have put a lot of former Communists back in power. A striking fact: former East Germans are markedly more socialist than other Germans, even though they saw the horrors of socialism first-hand.

How is this possible? There are many answers, most of them complementary. Check out my paper if you're curious.

TM: In your book, you "triangulate" different points of view regarding democratic outcomes. There is the classic pro-democracy (somewhat anti-market) crowd, the classic public choice (assuming voter rationality) crowd, and then your thesis (assuming voter irrationality). Much of the online commentary has come from supporters of the first view and last view. What criticisms have you received from the classic Public Choice crowd? Do you see them incorporating your model anytime soon? Will Gordon Tullock even speak to you anymore?

BC: Last question first: Last week Gordon popped in my office and asked to see my book. I gave him one of my last copies – how could I refuse? Now if you don't know Gordon, let me tell you: his highest praise is normally "This is better than I thought you were capable of." But within hours, Gordon returned to my office to praise the book. True story.

From other people in Public Choice, the main objection has been semantic: "Say ignorant, not irrational." I spent several pages trying to change their minds, but I don't know if I've succeeded.

TM: There is an idea in your book about political "slack", the "wiggle room" that politicians and bureaucrats have to go against public opinion for the public's own good. Would it be possible for an extreme supporter of democracy to argue that this slack could be used to overcome the public's irrationality, and thus, achieve rational policies? Is it likely?

BC: "Could be"? Yes. "Does"? That's a hard argument to make – there are too many bad policies right in front of our eyes.

Still, I freely admit that, for various reasons, democracy delivers better policies than the median citizen wants. A defender of democracy could and should focus on these reasons – and figure out better ways to leverage them.

TM: And how does this idea of slack (which you seem to characterize to move policies for the better) jibe with "The Idea Trap", which would argue for the movement in the opposite direction?

BC: I'd say these two ideas are unrelated – one is a static claim, the other is about social dynamics. The first claim says that due to slack, policies are better than the public wants. The second claim says that low (or negative) economic growth makes bad policies more popular, which reinforces low growth.

TM: If you were forced to write another chapter (or expand on a current chapter), what would it be? What are going to be the "deleted scenes" when the "director's cut" comes out?

BC: If I had really good data on the connection between public opinion and policy, I'd write another chapter about that connection to tighten up the argument – and test for the importance of different kinds of slack.

Before I submitted the book to Princeton, I deliberately cut 10% of the words from every chapter to raise the quality of the writing. What went out? Nothing I really miss. As a writer, I truly believe that less is more.

TM: Has anything come out since the publishing of the book that would add to or modify your theses?

BC: It's only been two months, so the honest answer is no. Give it time.

TM: Is there any evidence in the literature that sheep behave rationally irrational?

BC: Funny. My colleague Robin Hanson saw the cover, and immediately responded: "It's too optimistic. Sheep could herd around the truth. You're saying that human beings herd around error." So I guess he's right – the cover is unfair to sheep.

TM: You make a strong case in your book to give experts the benefit of the doubt unless there is evidence to the contrary. But there are certainly current and historical examples of experts getting things very wrong. From where you sit, are there specific conditions that predispose to expert error?

BC: A lot of expert error stems from experts' desire to not stray too far from the conventional beliefs of their society. No one – even an expert - likes being a pariah. For example, I suspect that those economists who oppose a free market in organs would change their minds if non-economists weren't horrified by the idea.

The other big problem is that experts, like other people, often get very attached to particular conclusions, and then compound the problem with social pressure. If things get bad enough, the field "turns into a religion." Religion itself is the most obvious example – the world's foremost experts on e.g. Catholicism are devout Catholics who feel strong psychological and social pressure to avoid hard questions.

At this point, I suspect someone is saying: "Aha! That's just what economics is like." All I can say to this person is: Come in and see for yourself. You may not like what you see, but you will find that economists are exceptionally reasonable people. When you disagree with the typical economist, you can expect measured argument, not hostility – even though economists address the kinds of questions that people really care about. You can have a civil conversation with most economists about practically anything; what other profession can you say that about?

TM: You don't stray far in the book from the economic sphere when talking about irrational beliefs of the public, with one brief dalliance into toxicology. Where else is there evidence you have come across of irrationally held beliefs with effects on policy?

BC: Many of the public's beliefs about foreign policy are quite irrational. A nice example: Public opinion researchers have documented a strong "rally-round-the-flag" effect. Declaring war makes people think that war is a good idea. Furthermore, as time goes on, support for war predictably declines, even if the war goes as well as could be reasonably suspected.

Similarly, as John Mueller argues in Overblown, the public vastly overestimates the danger of terrorism and the value of "doing something" about it – biases that have dominated U.S. policy for the last six years.

TM: If more poorly educated voters lead to more irrational economic policies, is this a reason to oppose a lot of immigration (presuming immigrants would be less educated)?

BC: This is a complicated question; I'll blog it in coming weeks. Short version:

If you care as much about immigrants as natives, this is no reason to oppose immigration. Consider the following example:

Suppose there are two countries with equal populations. The quality of policy ranges from 0-10, 10 being best. In country A, bliss points (people's first choice for policy) are uniformly distributed from 2-6. In country B, bliss points are uniformly distributed from 4-8.

What does democratic competition deliver? When the countries are independent, country A gets a policy quality of 4 (the median of the uniform distribution from 2-6), and country B gets a policy quality of 6 (the median of the uniform distribution from 4-8). Average policy that people live under: 50%*4+50%*6=5.

Now suppose you open the borders, and everyone moves to country B (the richer country). The median of the whole distribution is 5. Result: The immigrants live under better policies, the natives live under worse policies. The average (5) remains unchanged.

That's small consolation to the natives, you say? Keep in mind that even in the simplest model, the economic benefits of immigration for natives could easily outweigh the political losses for natives.

In any case, the simplest model seriously overstates natives' political losses for natives. Some of the main reasons:

• Empirically, non-natives are markedly less likely to vote than natives, even controlling for education and age. Immigration has a considerably smaller effect on the median voter than it does on the median resident.
• Natives start with a near-monopoly on political slack. At least initially, all of the incumbent politicians, government officials, media leaders, etc. will be natives, and will tend to use their slack to prevent deterioration of the political status quo.
• "Faith in rulers," another source of political slack that I discuss in my book, makes immigrants more likely to simply accept whatever policies are already in place.
• Although poor immigrants are likely to support a bigger welfare state than natives do, the presence of poor immigrants makes natives turn against the welfare state. Why would this be? As a rule, people are happy to vote to "take care of their own"; that's what the welfare state is all about. So when the poor are culturally very similar to the rich, as they are in places like Denmark and Sweden, support for the welfare state tends to be uniformly strong.

As the poor become more culturally distant from the rich, however, support for the welfare state becomes weaker and less uniform. There is good evidence, for example, that support for the welfare state is weaker in the U.S. than in Europe because our poor are disproportionately black. Since white Americans don't identify with black Americans to the same degree that rich Danes identify with poor Danes, most Americans are comfortable having a relatively small welfare state.

Thus, even though black Americans are unusually supportive of the welfare state, it is entirely possible that the presence of black Americans has on net made our welfare state smaller by eroding white support for it.

Immigration is likely to have an even stronger counter-balancing effect on natives' policy preferences because, as far as most Americans are concerned, immigrants from Latin American are much more of an "out-group" than American blacks. Faced with the choice to either cut social services or give "a bunch of foreigners" equal access, natives will lean in the direction of cuts. In fact, I can't think of anything more likely to make natives turn against the welfare state than forcing them to choose between (a) helping no one, and (b) helping everyone regardless of national origin.

Finally, let me add that even if immigrants do have negative political consequences, solutions should focus on the alleged problem. If you don't like how immigrants vote, the solution is to deny immigrants the right to vote. If you don't want to pay taxes to support immigrants, the solution is to make immigrants ineligible for benefits. Whatever you do, don't campaign to close the border and deport millions of people back to Third World poverty.

TM: In my area of work, health care, people exhibit a plethora of irrational behavior - "alternative medicine" and snake-oil sales being as profitable as it is. Is this evidence against "rational irrationality", or can it be explained?

BC: My basic claim is just that demand for irrationality has the standard negative slope. If consumers are less than fully rational about health care, but are even worse as voters, it is evidence in favor of my approach. However, if people are more rational about health care as voters than as consumers, that would be evidence against my approach.

TM (via my co-blogger Matt McIntosh): There are a couple of stories in the book that you like to tell: The first is about how when you ask people about their motives in voting and look at their actual voting record, there isn't generally a consistent pattern of self-interest being expressed. Yet when you ask them about other people's motives in voting, they often attribute self-interested motives to their choices. The second story is about how everyone seems to think that there's collusion and other such shady business dealings in every industry except the one they're familiar with, which is honest work full of cutthroat competition.

This is interesting because it goes contrary to what one might expect: More commonly you see people over generalizing from their own experiences because these are the kind they're familiar with. Instead here you seem to see people doing the opposite. Why does this "cynical bias" occur?

BC: Couldn't this is just another example of what I call pessimistic bias? After all, you know your own motives, but have to guess at those of others. So when we guess about others' motives, our tendency to pessimism makes our guess unduly negative.

TM: In many reviews and comments from prominent people, one word keeps coming up with regard to you contribution to the field: important. How do you feel about this and do you think your work could be a turning point in the way scholars approach these issues?

BC: I'd really like to believe this - so I need to try extra hard to be realistic. The hard truth is that there is a lot of inertia in academia, and economics revolves around journal articles, not books. At the same time, Psychology and Economics has been exerting pressure on political economy to face facts for a decade or so. On balance, then, I'd say there's a 10% chance that my book is influential enough to significantly change the way that economists think about democracy. I wish it were higher, but I've still sharply exceeded my own expectations.