Friday, December 07, 2012
BOB GARFIELD: Last week the popular prediction market Intrade announced it would shut its doors to Americans, after being sued by US regulators. A prediction market is where people bet on the likelihood of any future event, such as will Syrian President Bashar al-Assad’s regime last another year, will NASA find extraterrestrial life by New Year's Eve, will “Les Mis” win Best Picture? US regulators have accused Intrade of violating the ban on off-exchange options trading or, putting it another way, gambling. But others argue that sites like Intrade can be better predictors than pundits, certainly, or even polls. Washington Bureau Chief David Leonhardt has written about the markets. David, welcome to OTM.
DAVID LEONHARDT: Thanks, it’s good to be here.
BOB GARFIELD: All right, how does it work?
DAVID LEONHARDT: Well, investors or bettors can bet on real world events, as you said. In the typical contract, you get $10 if your bet comes true and nothing if it doesn’t. So you could bet on something like, a few months ago, will Barack Obama win the state of Ohio? And if the contract had been trading at, at 55%, that would mean essentially that you’re betting $5.50 and if he wins you get $10, and if he doesn't you get nothing. The advantage of these prediction markets is predictions have real value. They help us figure out how we should organize our lives. Predictions about the economy and about politics, they allow us to react and to plan. And if prediction markets are able to summon what is sometimes called “the wisdom of crowds” then they really do have value, even to people who are not playing in those markets.
BOB GARFIELD: We had your colleague Nate Silver on a little while ago [LAUGHS] and he made very clear how terrible the pundits are at predicting the future. One of the virtues of these prediction markets is that, with a few notable exceptions, they are very good.
DAVID LEONHARDT: They have a good record. It is not perfect, and I think the main flaws that you see in it stem from two things. One, uncertainty, and two, the markets have been too small. If you and I just decided to bet on the election, we wouldn't be that great about it, necessarily, but if it was you and I and 100,000 of our closest friends, the biases would tend to be cancelled out and you would have information from all over the place. And so, it is true when you look back on, on past elections, 2006 and 2008, these prediction markets did a really nice job of synthesizing information about who was really likely to win.
BOB GARFIELD: Give me an example of how this sleuthing can play out.
DAVID LEONHARDT: Imagine it’s Election Night and you are someone who plays on Intrade, and you happen to be following the vote tallies as they come in from a particular county in Indiana that you have decided is a meaningful county. Or, to even go one step further, imagine you, yourself go to local election board and try to pick up information before it’s posted on the website, and then you decide, you open up your laptop and you immediately start trading on that information and you get ahead of the curve. And then you buy up shares, but then also, as time goes on, reports start to come over the transom that Obama is looking better in Indiana, and people who don't have that information begin to make the same bet that you make and the price begins to move. And then the networks project that Obama will beat McCain in Indiana, and the market moves sharply to 98 or 99% that Obama would win. That’s how you could see this kind of coming together of private or expert information with public information.
BOB GARFIELD: Now, it’s clear that there is some social benefit to having accurate predictions about any number of future events. But to the government, since money is being wagered and money is being lost, you know, if it looks like a duck and it quacks like a duck, it's a duck, and it's a gambling duck, and they want to shut it down. If you had to go before the Commodities Future Trading Commission and advocate for predictions markets as being something other than gambling sites, what would you say?
DAVID LEONHARDT: I don’t have a view about this particular case, but broadly what I would say is once we've decided that as an economy and a society we are comfortable with lotteries, we are comfortable with Las Vegas and Atlantic City casinos, I have a hard time seeing the argument that we shouldn't then also be comfortable with prediction markets. They provide more social value than a blackjack game or than the lottery because they give us information about the future that we can then use. And you can say, hah, the prediction market thinks that this candidate is favored to win maybe two to one, and that allows me to think about how I want to plan my business, based on what I think’s gonna happen with taxes or with regulation. And so, it really does have a use.
BOB GARFIELD: Well, since we’re talking about predictions, we now have the federal government trying to move against one particular predictions market. What’s gonna happen in this area in the future?
DAVID LEONHARDT: In the end, we will have functioning prediction markets. I think prediction markets are too similar to too many other things that we have in our economy – lotteries, gambling, stock markets - and are too useful and are made too easy by the Internet not to be with us eventually. So I would predict with maybe 80% confidence that within ten years we will have bigger prediction markets than we have today and they will be a normal part of our dialogue.
BOB GARFIELD: David, thank you.
DAVID LEONHARDT: Thank you.
BOB GARFIELD: David Leonhardt is the Washington Bureau chief for the New York Times.
[MUSIC/MUSIC UP AND UNDER]