Q&A With Elections Expert David Schleicher

David Schleicher, a law professor at George Mason University, talks about rational ignorance, partisan branding and why voters don't know much about politics.

Explain what you mean by rational ignorance.

Voters have very little reason or incentive to learn very much about politics. It doesn't affect your day-to-day life. The odds that your vote is going to matter in an election are very small. This is not to say that voters are incapable of learning about politics. It's just that it doesn't make a lot of sense to become very informed about politics. In most elections, the way that this ignorance is overcome, or at least personally overcome, is through partisan information. So although we don't know very much about the intricacies of legislation, we're able to use the partisan label of the people who are in office. ... So if Barack Obama is president, and we know he's a Democrat, and someone likes or doesn't like something the government is doing, they're able to add this to a tally of their information about whether they're a Democrat or a Republican.

So party ID becomes kind of a shorthand.

There's an enormous amount of debate among political scientists about how far using the shorthand gets you—whether it gets you fully informed or partially informed. But it's certain that, particularly outside of big races, when we're not talking about the presidency, partisan information counts for almost everything we know about politics. Almost no one can name their councilman if they're in a big city. And if they can name them, they can't say much about their policy preferences. But party carries a lot of information, at least in national elections. I think that it carries less information in local elections.

Why is that?

Our preferences in national elections and local elections don't necessarily track one another. There's a lot of evidence that people's preferences about whether they think there should be more or less development are not closely related to whether they think the federal government should be spending more or whether we should go to war in Iraq. ... It doesn't carry zero information. ... In partisan elections, people will almost always vote for candidate from the party of the candidate they supported in the presidential election.

There's a coattails effect.

It's extremely dramatic in local elections, but it's not always true. In New York City, despite the fact that people support Democrats for president, they sometimes break away from this and vote for a Republican for mayor. But the thing is, people are able to get some information from party labels. In nonpartisan elections, it's basically taking a situation where people have little information and reducing it to zero. ... Nonpartisan elections show extreme fall-off in turnout, and there are extremely high reports of ignorance in a variety of ways. In the absence of party information, people rely on ethnic cues to a greater degree than they do in partisan elections.

Does gender also play a role?

Gender is a powerful factor in nonpartisan city elections. In partisan elections, most people are just going to vote the party. And that is not as good as it would be in national elections, where it is a close indicator of what the politicians you are voting for are going to do. But it's better than nothing. If you know a Tucson candidate is a Democrat or a Republican, you know something about them, whereas in nonpartisan elections, you don't know anything at all.

It seems as though it's when local government does something that really affects the average voter thatthey sit up and pay attention—if they institute a new tax, for example.

And that's the real problem with nonpartisan elections. When someone does something you like or don't like on the national level, you say, "A Republican did that," or, "A Democrat did that." ... When someone in a nonpartisan system does something, you have no idea who to credit or blame.

Is it unreasonable for voters to not pay attention to politics?

Politics is so complicated that unless you get paid to cover it, figuring out all the intricacies is going to be impossible.

You have to have some kind of financial interest in it, like a government consultant.

Or you have to be a cranky old person.

Or it's your hobby. There are people who love volunteering on campaigns or reading about politics.

It's a consumptive good. It's how I am about politics. I love politics. But it's not because I think I'm going to get some kind of personal gain out of it. People are busy. ... The broad point is: The way elections systems should work is that they should take into account that voters have problems learning about politics or limits in how much they know about politics. (Systems) should do their best to give them the best information they can. ... Parties are brands, and we can give them credit. It allows us to overcome the lack of information we have about politics.

You've noted that turnout tends to drop in nonpartisan elections. I'm looking at turnout in Tucson's elections, and it's pretty low—about 25 percent some of the time.

One of the things you ask, when you're deciding whether to vote, is how costly casting even a moderately informed vote is going to be to you. You're more likely to vote if you care about who wins. One of the things that will make you care is whether you know anything, and so the cost of information is related to whether you vote. If you increase the cost of information, which nonpartisan elections are going to do by removing information, you increase the cost of voting, which decreases the supply.

So even though we have bad turnout on our off-year elections, it could get worse if we went to a nonpartisan system.

Absolutely. Nonpartisan elections were kind of a progressive innovation at the beginning of the 20th century that were originally intended to limit the ability of machine parties to mobilize their voters. Nonpartisan elections are probably the worst electoral innovation of the 20th century, because they removed so much information from politics.