BROOKE GLADSTONE: According to an analysis by marketing firm IHS Global Insight, the government shutdown will cost us about $300 million dollars a day in lost economic output. So, if you wanted, you could call that the cost of political intransigence in the nation's capital. But how much would it cost to make we, the people less mulish? Well, according to a study published this summer by the National Bureau of Economic Research, you could pay for it with an Amazon gift card. That's right. Researchers divided volunteers into two groups and asked them basic factual questions about politics. The participants in one group were told that their odds of winning the card increased with accuracy. The other group was just ask the questions, no gift card was mentioned. The possibility of tangible reward boosted accuracy profoundly, says Gregory Huber, who was one of the researchers.
GREGORY HUBER: We see divides between Democrats and Republicans on very simple factual matters: How old was Barack Obama when he ran for president? How old was John McCain when he ran for president? But when you pay people, you find that the size of those differences between Democrats and Republicans is substantially smaller.
BROOKE GLADSTONE: How much smaller?
GREGORY HUBER: About half of the gap between Democrats and Republicans went away.
BROOKE GLADSTONE: [LAUGHS] Did you ever deduct points for getting an answer wrong?
GREGORY HUBER: Well, that was the spirit of the second experiment, where in addition to paying people to get the answer right, we said you could get some portion of that amount for simply saying I don't know. And, in that context, we found that the gaps between Democrats and Republicans were about 80 percent smaller –
- [LAUGHS] than when you didn't pay them.
BROOKE GLADSTONE: Can you give me an example of questions that had more political freight?
GREGORY HUBER: The question about the unemployment rate during the first three years of the Obama administration. In the condition where we didn't pay people, we found that the gap was about 30 points. And when we paid people for the correct response, it was only about 10 points.
BROOKE GLADSTONE: [LAUGHS] So what does this mean, that lacking the appropriate incentive, people are more likely to lie to themselves?
GREGORY HUBER: Well, I think it might mean that when people are answering surveys, they're using the survey not as a chance to try to tell you what they really think, but as a chance to cheerlead for their partisan team, because if this ends up in the newspaper and it turns out Republicans don't like a Republican policy, that will be used against Republicans.
BROOKE GLADSTONE: Therein lies the problem with applying your findings, because you're bringing people into an experiment where their opinions aren't going to be publicly expressed in a poll, so it's a lot easier to abandon your team for five bucks.
GREGORY HUBER: I think that's exactly right. We have an intuitive sense that when people are answering surveys they aren't necessarily taking themselves too seriously, but we’ve had a hard time figuring out how large that problem is. And these data suggest that it's perhaps very large. So when you hear extensive reports about Democrats and Republicans having stark disagreements about even the basic facts, maybe we shouldn't be too worried because maybe they know that they don't really disagree, but the survey data aren’t going to allow us to tell that.
BROOKE GLADSTONE: So is there any way to put this idea into our drinking water? I mean, what can you do with it?
GREGORY HUBER: What to do with it, maybe we should be skeptical that people really believe those responses that they give to a media survey. I think it tells us about what those polls really mean deep down inside, because if at the end of the day we thought that Democrats and Republicans looked back at the past and misremembered the facts, we’d be very, very worried about democracy. Instead, it appears that they, to a large degree, understand what really happened. They might still prefer Democrats to Republicans, but it's not because they can't remember what happened or they actually doubt the reality.