Read an Excerpt

THE MYTH OF THE RATIONAL VOTER

PRINCETON UNIVERSITY PRESS

Chapter One

I am suspicious of all the things that the average citizen believes. -H. L. Mencken, A Second Mencken Chrestomathy

What voters don't know would fill a university library. In the last few decades, economists who study politics have revitalized age-old worries about the people's competence to govern by pointing out that-selfishly speaking-voters are not making a mistake. One vote has so small a probability of affecting electoral outcomes that a realistic egoist pays no attention to politics; he chooses to be, in economic jargon, rationally ignorant.

For those who worship at the temple of democracy, this economic argument adds insult to injury. It is bad enough that voters happen to know so little. It remains bearable, though, as long as the electorate's ignorance is a passing phase. Pundits often blame citizens' apathy on an elections' exceptionally insipid candidates. Deeper thinkers, who notice that the apathy persists year after year, blame voters' ignorance on lack of democracyitself. Robert Kuttner spells out one version of the story:

The essence of political democracy-the franchise-has eroded, as voting and face-to-face politics give way to campaign-finance plutocracy ... [T]here is a direct connection between the domination of politics by special interest money, paid attack ads, strategies driven by polling and focus groups-and the desertion of citizens. ... People conclude that politics is something that excludes them.

Yet the slogan "The solution for the problems of democracy is more democracy" sounds hollow after you digest the idea of rational ignorance. Voter ignorance is a product of natural human selfishness, not a transient cultural aberration. It is hard to see how initiatives, or campaign finance reform, or any of the popular ways to "fix democracy" strengthen voters' incentive to inform themselves.

As the rational ignorance insight spread, it became an intellectual fault line in the social sciences. Economists, along with economically minded political scientists and law professors, are generally on one side of the fault line. They see voter ignorance as a serious problem, making them skeptical about using government intervention to improve market outcomes. Beneficial government action is possible in theory, but how could hopelessly uninformed voters be expected to elect politicians who follow through? The implication: "Voters don't know what they're doing; just leave it to the market." Thinkers on the other side of the fault line downplay these doubts about government intervention. Once you discount the problem of voter ignorance, it is a short hop from "the policies beneficial in theory" to "the policies democracies adopt in practice."

In time, rational ignorance spawned an expansive research program, known as public choice or political economy or rational choice theory. In the 1960s, finding fault with democracy bordered on heretical, but the approach was hardy enough to take root. Critiques of foolish government policies multiplied during the 1970s, paving the way for deregulation and privatization.

But as these ideas started to change the world, serious challenges to their intellectual foundations surfaced. Earlier criticism often came from thinkers with little understanding of, and less sympathy for, the economic way of thinking. The new doubts were framed in clear economic logic.

The Miracle of Aggregation

Think about what happens if you ask a hundred people to run a 100-meter race, and then average their times. The average time will not be better than the time of the fastest runners. It will be worse.... But ask a hundred people to answer a question or solve a problem, and the average answer will often be at least as good as the answer of the smartest member. With most things, the average is mediocrity. With decision-making, it's often excellence. You could say it's as if we've been programmed to be collectively smart. -James Surowiecki, The Wisdom of Crowds

If a person has no idea how to get to his destination, he can hardly expect to reach it. He might get lucky, but common sense recognizes a tight connection between knowing what you are doing and successfully doing it. Ubiquitous voter ignorance seems to imply, then, that democracy works poorly. The people ultimately in charge-the voters-are doing brain surgery while unable to pass basic anatomy.

There are many sophisticated attempts to spoil this analogy, but the most profound is that democracy can function well under almost any magnitude of voter ignorance. How? Assume that voters do not make systematic errors. Though they err constantly, their errors are random. If voters face a blind choice between X and Y, knowing nothing about them, they are equally likely to choose either.

What happens? With 100% voter ignorance, matters are predictably grim. One candidate could be the Unabomber, plotting to shut down civilization. If voters choose randomly, the Unabomber wins half the time. True, the assumption of zero voter knowledge is overly pessimistic; informed voters are rare, but they do exist. But this seems a small consolation. 100% ignorance leads to disaster. Can 99% ignorance be significantly better?

The surprising answer is yes. The negative effects of voter ignorance are not linear. Democracy with 99% ignorance looks a lot more like democracy with full information than democracy with total ignorance. Why? First, imagine an electorate where 100% of all voters are well informed. Who wins the election? Trivially, whoever has the support of a majority of the well informed. Next, switch to the case where only 1% of voters are well informed. The other 99% are so thick that they vote at random. Quiz a person waiting to vote, and you are almost sure to conclude, with alarm, that he has no idea what he is doing. Nevertheless, it is basic statistics that-in a large electorate-each candidate gets about half of the random votes. Both candidates can bank on roughly a 49.5% share. Yet that is not enough to win. For that, they must focus all their energies on the one well-informed person in a hundred, who takes the prize? Whoever has the support of a majority of the well informed. The lesson, as Page and Shapiro emphasize, is that studying the average voter is misleading:

Even if individuals' responses to opinion surveys are partly random, full of measurement error, and unstable, when aggregated into a collective response-for example, the percentage of people who say they favor a particular policy-the collective response may be quite meaningful and stable.

Suppose a politician takes a large bribe from "big tobacco" to thumb his nose at unanimous demand for more regulation. Pro-tobacco moves do not hurt the candidate's standing among the ignorant-they scarcely know his name, much less how he voted. But his share of the informed vote plummets. Things get more complex when the number of issues rises, but the key to success stays the same: Persuade a majority of the well informed to support you.

This result has been aptly named the "Miracle of Aggregation." It reads like an alchemist's recipe: Mix 99 parts folly with 1 part wisdom to get a compound as good as unadulterated wisdom. An almost completely ignorant electorate makes the same decision as a fully informed electorate-lead into gold, indeed!

It is tempting to call this "voodoo politics," or quip, as H. L. Mencken did, that "democracy is a pathetic belief in the collective wisdom of individual ignorance." But there is nothing magical or pathetic about it. James Surowiecki documents many instances where the Miracle of Aggregation-or something akin to it-works as advertised. In a contest to guess the weight of an ox, the average of 787 guesses was off by a single pound. On Who Wants to Be a Millionaire, the answer most popular with studio audiences was correct 91% of the time. Financial markets-which aggregate the guesses of large numbers of people-often predict events better than leading experts. Betting odds are excellent predictors of the outcomes of everything from sporting events to elections. In each case, as Page and Shapiro explain, the same logic applies:

This is just an example of the law of large numbers. Under the right conditions, individual measurement errors will be independently random and will tend to cancel each other out. Errors in one direction will tend to offset errors in the opposite direction.

When defenders of democracy first encounter rational ignorance, they generally grant that severe voter ignorance would hobble government by the people. Their instinctive responses are to (a) deny that voters are disturbingly ignorant, or (b) interpret voters' ignorance as a fragile, temporary condition. To call these responses "empirically vulnerable" is charitable. Decades of research show they are plain wrong. About half of Americans do not know that each state has two senators, and three-quarters do not know the length of their terms. About 70% can say which party controls the House, and 60% which party controls the Senate. Over half cannot name their congressman, and 40% cannot name either of their senators. Slightly lower percentages know their representatives' party affiliations. Furthermore, these low knowledge levels have been stable since the dawn of polling, and international comparisons reveal Americans' overall political knowledge to be no more than moderately below average.

You could insist that none of this information is relevant. Perhaps voters have holistic insight that defies measurement. But this is a desperate route for a defender of democracy to take. The Miracle of Aggregation provides a more secure foundation for democracy. It lets people believe in empirical evidence and democracy at the same time.

The original arguments about rational ignorance took time to spread, but eventually became conventional wisdom. The Miracle of Aggregation is currently in the middle of a similar diffusion process. Some have yet to hear of the Miracle. Backward-looking thinkers hope that if they ignore the objection, it will go away. But the logic is too compelling. Unless someone uncovers a flaw in the Miracle, the fault line in the social sciences will close. Economists and economically minded political scientists and law professors will rethink their doubts about democracy, and go back to the prerational ignorance presumption that if democracies do X, X is a good idea.

The Reality of Systematic Error

Universal suffrage, which to-day excludes free trade from the United States, would certainly have prohibited the spinning-jenny and the power-loom. beliefs about many other subjects. But as far as economics is concerned, the jury is in. People do not understand the "invisible hand" of the market, its ability to harmonize private greed and the public interest. I call this antimarket bias. People underestimate the benefits of interaction with foreigners. I call this antiforeign bias. People equate prosperity not with production, but with employment. I call this make-work bias. Lastly, people are overly prone to think that economic conditions are bad and getting worse. I call this pessimistic bias.

Economic policy is the primary activity of the modern state, making voter beliefs about economics among the most-if not the most-politically relevant beliefs. If voters base their policy preferences on deeply mistaken models of the economy, government is likely to perform its bread-and-butter function poorly. To see this, suppose that two candidates compete by taking positions on the degree of protectionism they favor. Random voter errors about the effect of protection cause some voters who prefer the effect of free trade to vote for protection. But it is equally common for voters who prefer the effect of protection to vote for free trade. Then the Miracle of Aggregation holds: in spite of voter ignorance, the winning platform is socially optimal.

For anyone who has taught international economics, though, this conclusion is underwhelming. It takes hours of patient instruction to show students the light of comparative advantage. After the final exam, there is a distressing rate of recidivism. Suppose we adopt the more realistic assumption that voters systematically overestimate the benefits of protection. What happens? Lots of people vote for protection who prefer the effect of free trade, but only a few vote for free trade who prefer the effect of protection. The political scales tilt out of balance; the winning platform is too protectionist. The median voter would be better off if he received less protection than he asked for. But competition impels politicians to heed what voters ask for, not what is best for them.

Comparable biases plausibly underlie policy after policy. For example, supply-and-demand says that above-market prices create unsaleable surpluses, but that has not stopped most of Europe from regulating labor markets into decades of depression-level unemployment. The most credible explanation is that the average voter sees no link between artificially high wages and unemployment. Before I studied economics, I failed to see it myself.

Modern Research versus Intellectual Tradition

Economists have two attitudes toward discourse, the official and the unofficial. -Donald McCloskey, The Rhetoric of Economics

The terminology of "systematic" versus "random" error entered economists' vocabulary about 30 years ago. But the concept of systematic error has a much longer history. Here is how Simon Newcomb began an article in the Quarterly Journal of Economics in 1893:

The fact that there is a wide divergence between many of the practical conclusions of economic science, as laid down by its professional exponents, and the thought of the public at large, as reflected in its current discussion and in legislation, is one with which all are familiar.

This was the intellectual climate that Newcomb saw in the contemporary United States and Great Britain. Over a century earlier, in The Wealth of Nations, Smith made similar observations about economic beliefs in Britain:

Nothing, however, can be more absurd than this whole doctrine of the balance of trade, upon which, not only these [mercantilist] restraints, but almost all other regulations of commerce are founded. When two places trade with one another, this doctrine supposes that, if the balance be even, neither of them loses or gains; but if it leans in any degree to one side, that one of them loses, and the other gains in proportion to its declension from the exact equilibrium.

(Continues...)

---

Excerpted from THE MYTH OF THE RATIONAL VOTER by Bryan Caplan Copyright © 2007 by Princeton University Press. Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

---