Jan. 8, 2008 at 8:47 PM ET
If you’re the speculating type, now’s the time to sell high on Obama and buy low on Clinton: The past few days have seen a dramatic shift in the online political markets, where traders put down money to predict which candidates will prevail in the primary season. Since last week’s Iowa caucus, prices for Democratic hopeful Barack Obama’s stock on Intrade have shot up from the 20s to the 70s, while rival Democrat Hillary Clinton has experienced a market crash of similar proportions. It’s enough to set economists wondering once more about how prediction markets work.
The reversal of fortunes for Obama and Clinton is reflected not only on Intrade, but also on the Iowa Electronic Markets - which pioneered the idea of using real-money trading to capture the "wisdom of crowds" and predict the outcome of events ranging from traditional market decisions to football games.
Ironically, the spark for the past week's big market move also came from Iowa, in the form of presidential caucus results that made Obama into the new frontrunner for the Democratic nomination.
Some economists have suggested that prediction markets do a better job than polling when it comes to predicting political outcomes, because participants are backing up their opinions with real dollars. But the past week demonstrates the flip side of the relationship: Real events can have a dramatic effect on where that real money is going.
It's a bit of a puzzle for Justin Wolfers, a Wharton School economist who has been studying prediction markets for years. "We should really infer a lot out of Iowa," he told me today. In today's column for The Wall Street Journal, Wolfers notes that a relatively small, unrepresentative number of early voters in Iowa and New Hampshire can have an inordinately large impact on market trends.
Academics as well as political operatives are intrigued by this "momentum effect," and speaking as an economist rather than a political spin doctor, Wolfers said there might be a couple of reasons behind it.
"One view would be that everyone wants to vote for a winner," he told me. "But a different view would be that the people in Iowa and New Hampshire actually get to meet the candidates."
That could turn the voters in those states into a cadre of more knowledgeable, more respected players in the market - something that reminded me of the role that financial analysts play in traditional market sectors. "That's an interesting way of thinking about it," Wolfers said.
Brown University economists Brian Knight and Nathan Schiff delve more deeply into the momentum effect in a working paper analyzing the 2004 Democratic primary campaign for the National Bureau of Economic Research. The researchers find that "early voters have up to 20 times the influence of late voters, demonstrating a significant departure from the ideal of 'one person, one vote.'"
University of Iowa economist Tom Rietz, a member of the Iowa Electronic Markets' steering committee, said the big role played by the early primaries isn't surprising. "As they resolve uncertainty, they'll move the market one way or the other," he told me.
One of his big research interests is figuring out exactly when the political prediction market solidifies into a consensus pick. "It's usually sometime in between Iowa and Super Tuesday or Mini-Tuesday," Rietz said. "We're trying to research that question as we go."
Such questions aren't just academic. For one thing, the political prediction markets are providing insights into bigger-ticket markets where there's an eventual all-or-nothing payout, such as the binary options on the Chicago Board of Trade. "There are some unusual price dynamics where a contract is worth either a dollar or zero," Rietz said.
For another thing, some people are putting real money on political markets, although that's not strictly required. Accounts on the Iowa Electronic Markets are limited to $500. Intrade works with play money as well as real money, and other prediction Web sites, such as NewsFutures and Foresight Exchange, are strictly play-money operations.
Wolfers said he has seen some situations where traders (or bettors) "invest" tens of thousands of dollars in their political prediction. "There are definitely people taking these markets seriously and putting down significant amounts of money," he said.
If you caught the Obama surge at just the right moment, you could have more than doubled your investment just over the past week. Or, if you were a Clinton investor, you would have seen more than half of your stake disappear.
The University of Iowa's Rietz said the Republican race is currently more interesting than the Democratic race, because no one candidate is dominant in the market. As voters finish going to the polls in New Hampshire, GOP hopeful John McCain appears to have the hot stock on Intrade, while Mitt Romney has slumped since Iowa.
So is now the time to invest in a slumping candidate, based on the buy-low, sell-high principle? Well, on one hand, there's always the chance for another reversal of fortunes. But on the other hand, the current pattern for the Democrats reminds Rietz of the 2004 campaign, when Howard Dean's stock slumped during the lead-up to what turned out to be a screamingly unsuccessful performance in the Iowa caucuses.
"After the caucus, Dean dropped off much further," Rietz recalled.
So don't expect me to be a financial adviser here - or a political adviser, for that matter. But do feel free to weigh in below with your own thoughts about the wisdom of political crowds.
Update for 10:15 p.m. ET: Now this is getting interesting: In after-hours trading on Intrade, Clinton's stock has risen from a closing price of 27 to 54, while Obama's shares have fallen from 70.7 to 43. John Edwards has dwindled from 2.6 to 0.9. In contrast, McCain's stock has held fairly steady, heading up just a bit from 34.2 to 36.3. Keep a watch on how the numbers rise and fall as the night goes on - it's clear these folks are watching the results unfold in real time.
Update for 12:55 p.m. ET Jan. 9: The University of Iowa says more than 23,000 shares were traded on the Democratic market on Tuesday, with almost half of those shares relating to Clinton's contracts. Right now Clinton is at 55.6, Obama at 41.0 and Edwards at 2.7. Those fluctuating numbers reflect the market's perception of the percentage chance that each of those candidates will win the Democratic nomination. If you were the speculating type (see above), you could have doubled your money in one day.
Update for 1:50 p.m. ET Jan. 9: The Wharton School's Justin Wolfers points to his morning-after column for The Wall Street Journal, in which he calls Clinton's win "one of the most surprising upsets in U.S. political history." Here's his bottom line:
"While Sen. Clinton's unexpected victory has yielded red faces among the punditocracy, this also provides a useful opportunity for emphasizing just what a prediction market forecast says. That the price of a contract paying $1 if Sen. Clinton won in New Hampshire was selling for seven cents doesn't suggest that she was a sure loser. Rather, these prices suggest a probabilistic statement that the ultimate outcome was about a 7 percent chance. And as any horseplayer can tell you, sometimes the long shots do win."