4/14/07. NPR broadcast an interview with Washington Post columnist Shankar Vedantam who has written about the "decoy effect" in politics -- the curious ways in which we compare candidates and decide who gets our vote. Did Nader spoil Gore chances in 2000? Or did he in fact help Gore?
NPR: It's still 9 months before any voters actually get to have a say about who'll get to be next year's presidential candidates. But, as we all know, the race for the nomination is already well underway. The latest horse-race polls show a lot of flux on the Republican side. The LA Times/Bloomberg poll has Senator John McCain, who used to be called the obvious front-runner, in third place now behind former New York mayor, Rudolph Giuliani, and former senator and actor, Fred Thompson, who hasn't even entered the race. Former Massachusetts governor Mitt Romney is behind them but doing well at fund-raising. The Democratic field seems to be a little clearer. Clinton, Obama, and then John Edwards, followed by several other candidates. That same poll shows Senator Hillary Clinton in first place, about ten points ahead of Senator Obama, But John Edwards' third-place showing is strong. And psychologists say that who's in third place can have a strong effect on how somebody may choose #1. Shankar Vedantam wrote about what psychologists call "the decoy effect" in his Washington Post column. Before we get into politics, let's talk about how the decoy effect works in other, maybe more mundane, areas of life. You write about, for example, a study by a marketing professor at Duke University who took a look at restaurant choices.
Shankar Vedantam: That's right. Joel Huber was a marketing professor at Duke University who gave this example -- and I'm going to change his numbers just a little bit to make it easier... Imagine you're choosing between two restaurants. One of them is a 5-star restaurant that's five miles away, and the other is a 3-star restaurant that's three miles away. Now, if the better restaurant was also closer, there would be no dilemma at all. But, as in many choices that we have to make in life, we have to trade off in this case quality against convenience. What Huber did is he introduced a third option to a group of people. He said, "What if you had a 4-star restaurant that was six miles away?" And when he presented that option, a lot of people said, "Well, the 5-star restaurant that's five miles away is both better than and closer than the new third candidate, so it's probably the best restaurant of all." But when Huber presented a different group of people with a different third option -- a 2-star restaurant that was four miles away -- now the 3-star restaurant started looking a lot better on both quality and convenience, whereas this 5-star restaurant was only better on quality! What's interesting about it was that the third candidate -- in this case a third restaurant -- is always inferior to either one of the front-running candidates. And so in a rational world, you would assume it would have no effect whatsoever on the decision. But it has a pretty big effect on which restaurant we end up choosing.
NPR: Let's chance nosing into politics with this. How could this ostensibly play out in either party?
SV: Well, politics is a little bit more complicated for several reasons. For one thing, you often have many more than just two or three candidates. Also, people are thinking of not just one or two dimensions -- such as quality and convenience -- but they're thinking of dozens of different issues. Your position on the war in Iraq. Your position on healthcare. Your position on taxes. And they are playing up all of these issues against one another in their heads. It's a very complicated decision. But what the decoy effect suggests is that when you have two front-runners -- and, as you pointed out, the Democratic field right now does have two front-runners -- Barack Obama and Hillary Clinton -- the characteristics that people perceive John Edwards bringing to the race can subtly alter whether they think whether Obama or Clinton is the better candidate.
NPR: How could the qualities that Senator Edwards projects -- in theory -- affect how people wind up feeling about Senator Obama or Senator Clinton?
SV: Before I give the example, I should also mention that the decoy effect works only if you're an undecided voter. Let's say you're an undecided centrist Democratic voter. You like Clinton on the fact that she's strong on national security. But you also want somebody who's a fresh face in Washington and you like Obama on that issue. So let's say you have Edwards enter the race and you see Edwards as more dovish than Obama on national security issues but also part of the same establishment as Clinton -- he was, after all, the vice presidential candidate in the last election. So Obama now looks better than Edwards on both counts, whereas Clinton beats Edwards on only the national security issue. And if you configure Edwards differently in the mind of this undecided voter, Clinton could end up looking much better than Obama. Which is why it could potentially stand to the advantage of both Clinton and Obama to draw attention to certain characteristics of Edwards, but not to others. But politics is a volatile game and you talk up a third candidate too much and the third candidate could stop being a third candidate and become a front-runner.
NPR: What about the Republican side? Are there any guesses?
SV: The Republican side is a little big more confusing, I think, because it's changing so rapidly. So I think it's safe to say the decoy effect is probably playing a role. But how exactly? I would be hesitant to make an assertion about that.
NPR: Do you think the decoy effect had any discernible outcome in the election of 2000?
SV: Well, before I reported this column, I would have said the presence of Nader in the race tipped the election to Bush, especially in Florida. Because some or many of those 9,000 votes for Nader would have gone to Gore if Nader had not been in the race. After reporting this column, I'm actually not entirely sure that's true. I think it's possible that Nader did indeed take away some votes from Gore -- people who were essentially to the left of Gore who found Nader to be a more appealing candidate. But I think it's also possible Nader ended up pulling some voters who were undecided between Gore and Bush to the Gore camp. And the way I would think about this is: imagine a straight line where you have Gore on the left and Bush on the right. And let's say you have an undecided voter who is somewhere in the middle between Gore and Bush. And now you have Nader enter the race. Nader is to the left of Gore. And what, essentially, the presidence of Nader does is it tips the overall scale so that Gore feels like the better candidate among the three candidates this voter is thinking about. It's possible that the presence of Nader in the race actually drew many voters to the Gore camp. Whether Nader ended up taking more votes away from Gore than he ended up giving to Gore -- that I don't think anyone really can answer.