Democracy is supposed to enact the will of the people. But what if the people have no clue what they’re doing?
- By Jason BrennanJason Brennan is the Robert J. and Elizabeth Flanagan family chair and associate professor of strategy, economics, ethics, and public policy at Georgetown University’s McDonough School of Business.
OK, so that just happened. Donald Trump always enjoyed massive support from uneducated, low-information white people. As Bloomberg Politics reported back in August, Hillary Clinton was enjoying a giant 25 percentage-point lead among college-educated voters going into the election. (Whether that trend held up remains to be seen.) In contrast, in the 2012 election, college-educated voters just barely favored Barack Obama over Mitt Romney. Last night we saw something historic: the dance of the dunces. Never have educated voters so uniformly rejected a candidate. But never before have the lesser-educated so uniformly supported a candidate. Trump supporters might retort: “That’s because Trump supports the little guy and Clinton helps the already privileged college grads.” But that’s false: Trump supporters in the primaries had an average income of about $72,000 per year. They aren’t rich, but make more than the national average and more than Clinton supporters.
Trump owes his victory to the uninformed. But it’s not just Trump. Political scientists have been studying what voters know and how they think for well over 65 years. The results are frightening. Voters generally know who the president is but not much else. They don’t know which party controls Congress, what Congress has done recently, whether the economy is getting better or worse (or by how much). In the 2000 U.S. presidential election, most voters knew Al Gore was more liberal than George W. Bush, but significantly less than half knew that Gore was more supportive of abortion rights, more supportive of welfare-state programs, favored a higher degree of aid to blacks, or was more supportive of environmental regulation.
Just why voters know so little is well-understood. It’s not that people are stupid. Rather, it’s that democracy creates bad incentives.
Consider: If you go to buy a car, you do your research. After all, if you make a smart choice, you reap the rewards; if you make a bad choice, you suffer the consequences. Over time, most people learn to become better consumers.
Not so with politics. How all of us vote, collectively, matters a great deal. But how any one of us votes does not. Imagine a college professor told her class of 210 million students, “Three months from now, we’ll have a final exam. You won’t get your own personal grade. Instead, I’ll average all of your grades together, and everyone will receive the same grade.” No one would bother to study, and the average grade would be an F.
That, in a nutshell, is how democracy works. Most voters are ignorant or misinformed because the costs to them of acquiring political information greatly exceed the potential benefits. They can afford to indulge silly, false, delusional beliefs — precisely because such beliefs cost them nothing. After all, the chances that any individual vote will decide the election is vanishingly small. As a result, individual voters tend to vote expressively, to show their commitment to their worldview and team. Voting is more like doing the wave at a sports game than it is like choosing policy.
The great political scientist Philip Converse once said: “The two simplest truths I know about the distribution of political information in moderate electorates are that the mean is low and variance is high.” In other words, most people know nothing, some know less than nothing (that is, they are systematically mistaken rather than just ignorant), and some know a great deal. In general, college-educated people are better-informed than those with a high school diploma, who are in turn better informed than those who did not finish high school.
When people hear such depressing statistics, they are quick to wag their fingers at America’s broken education system. “This just shows we need better teachers!” they cry. It’s a plausible argument, but wrong.
In fact, average Americans have completed more schooling now than 60 years ago, but they’ve remained equally ignorant about politics even as their education levels rose. More fundamentally, to blame the schools is to misunderstand why citizens know so little. The schools teach them most of what they need to know to vote well. But they forget it because the information is not useful. And the reason it is not useful is because their individual votes make no difference.
Others say the problem could be fixed by encouraging citizens to deliberate together. They believe getting random groups of Americans together to talk about politics will cause them to resolve their differences, become informed, and reach agreement. However, political scientists have been conducting large number of experiments testing how deliberation works. Even though the researchers in question almost always want deliberation to “fix” democracy, in general, they tend to find that it makes things worse, not better.
None of this would matter if political information had no effect on how citizens vote. But, in fact, it does. Every other year, the American National Election Studies survey 1) what voters know, 2) what policies they support, and 3) who they are (e.g., white or black, poor or rich, employed or not). With these three sets of data, it is then possible to determine how information, by itself, changes what voters want, because we can control for whatever effect race, gender, and income have.
Trump supporters might be upset to learn that this method reveals that high-information voters (regardless of their income, race, employment status, gender, or where they live) tend to favor free trade and are pro-immigration. It’s not just that Trump’s anti-trade and anti-immigrant agenda flies against the consensus of economists on the left, right, and center, but it’s precisely the platform informed voters reject — regardless of their backgrounds.
That’s not to say that high-information voters tend to favor the Democrats’ politics. In fact, high-information voters tend to have policy preferences that cut across party lines. For instance, high-information voters are pro-free trade, pro-immigration, in favor of criminal justice reform, wish to raise taxes to offset the deficit, anti-war, pro-gay rights, and skeptical that the welfare state can solve all our problems.
The real worry, though, is that when we look at the policy platforms of the two major parties, we see that both the Republicans and Democrats push agendas that tend to appeal to the uniformed and disinterested. We can’t quite blame them for that. After all, politicians need to win elections, and to do so, they have to appeal to voters. In a modern democracy, the uninformed will always greatly outnumber the informed. The quality of our candidates reflects the quality of our electorate. But democracy encourages our electorate to be bad quality.
There is no real solution to the problem of political ignorance, unless we are willing to break with democratic politics. Some economists, such as Robin Hanson, favor using specialized betting markets to choose policies. Law professor Ilya Somin favors radically decentralized federal systems that encourage citizens to vote with their feet. In my recent book Against Democracy, I discuss how we might experiment with epistocracy — where political power is widespread, as in a democracy, but votes are in some way weighted according to basic political knowledge. Most of these proposals set off alarm bells (usually among people who have not bothered to think carefully about how these systems work). But each proposal at least takes seriously that universal suffrage and voter ignorance go hand in hand.
Trump’s victory is the victory of the uninformed. But, to be fair, Clinton’s victory would also have been. Democracy is the rule of the people, but the people are in many ways unfit to rule.
Photo credit: John Sommers II/Getty Images