Living as we do in the land of the Giants, it seemed until tonight as if nothing were more important than the outcome of the World Series. Now that the Giants have won the Series, fans can turn their attention to the elections. Will Jerry Brown beat Meg Whitman for Governor of California? Will Carly Fiorina manage to unseat Barbara Boxer in the Senate race? And nationwide, which party will take control of the House? Tuesday's election will decide these questions and many others.
But perhaps you're not political at all, and care more about an entirely different contest: Brandy versus Jennifer Grey, the frontrunners in this season's Dancing with the Stars. You won't learn the answer for several more weeks--but as with the elections, you can place your bets now. As you'll soon see, these bets are revealing.
It's obviously too late to bet on this year's World Series, but let's pretend it's still Game 2. You're in a sports bar watching the game, and the guy next to you eggs you on about who will take home the series trophy. For whatever reason, you're convinced the Rangers are going to win. Suddenly the guy who's been talking to you puts a hundred-dollar bill on the table and says, “Why don’t we make this interesting? I'll give you 2 to 1 odds.” Does he know something that you don’t?
Chances are, he does--and we're not saying that because we already know the outcome. Economic studies have shown that if you let people place bets, then with the right rules, they'll often reveal what they know even if they'd never tell you otherwise. People who know that your team’s pitcher is sick, for example, are more likely to bet against the team and give better odds when doing so. The reason is simple: betting (and the attendant prospect of a bigger win) gives them an economic incentive to tell you the truth.
This is true not just for baseball and, in fact, not just in sports. Economists are famously skeptical of what people say, but people's actions—their "revealed preferences," in econ jargon—tell you what they really think. The best way to learn how much someone prefers free-range chicken over the farm-raised variety isn't to ask them in a focus group, but to see which they actually buy and for how much. The best way to predict how people will vote in the midterm elections isn't to take a poll, which can be biased in all sorts of ways, but to see who's ahead in a real-money prediction market, like the ones run by the University of Iowa. Unlike polls, the Iowa prediction markets let people bet their money on candidates, so the most informed players drive up the prices of those candidates who are most likely to win. If you have information that others don't—whether it be about Meg Whitman's prospects in the California governor's race or Jennifer Grey's chances of winning this season's Dancing with the Stars—you'll have reason to make a bet and correct the errors of the other participants. In this way, the market prices can capture all the information ordinarily residing in people's heads.
This principle extends beyond prediction markets or betting pools. Consider the more prosaic "advance purchase order," which in the workaday world of business helps forecast actual orders. How? Imagine you're making laptops and you sell them to end consumers through Best Buy. By giving Best Buy a price break for ordering laptops three months in advance, you give Best Buy the incentive to forecast accurately, and pass that forecast to you through the “advance purchase order.” This is a more subtle way to let Best Buy bet their money on the future—in this case, how many of your laptops they can sell.
Insurance companies are geniuses at getting you to put your money where your mouth is. They can elicit your private information about your risk of getting sick even without looking at your medical history. Your insurer has a good idea of how likely you are to file a claim depending on whether you choose a policy with a high deductible (and low premiums) or vice-versa. Your cell-phone service provider puts the same principle to good use, too, by giving you a choice between a fixed monthly fee and a pay-as-you-go plan. Your choice reveals the kind of customer you are. The company wins either way because the information is useful for planning service capacity, such as how many new towers to build.
None of this is to say that implementing these principles is a snap. A prediction market, for example, is prone to occasional bubbles just as the stock market is. And figuring out how to make inferences from people's choices usually means understanding something about their decision-making behavior (such as risk attitude), which inconveniently varies from person to person. But even so, the idea of using some form of betting to elicit information is greatly underused in business and in politics. The world is filled with uncertainty, yet everybody knows a few things with some degree of certainty, especially about themselves. Can you think of ways to use this insight in your own business?
Copyright Kay-Yut Chen and Marina Krakovsky,http://www.secretsofthemoneylab.com
Chen and Krakovsky are co-authors of Secrets of the Moneylab: How Behavioral Economics Can Improve Your Business (Portfolio/Penguin, 2010).
