May 15, 2008 at 8:22 PM ETPrediction markets have been known to outdo the pollsters when it comes to handicapping political campaigns, and they can also be used to predict how bad the next flu epidemic can get, how well the next product will sell - or even how long the latest celebrity marriage will last. (Are you listening, Mariah Carey?)
But are these markets legit? That’s what researchers and regulators want to find out.
Most recently, the Iowa Electronic Markets - the country's oldest political prediction market working with real money for research purposes - foresaw that Barack Obama was gaining a lock on the Democratic presidential nomination long before the pundits came around.
Iowa Electronic Markets
|This chart shows the rise and fall of Democratic |
presidential candidates' fortunes on the Iowa
Electronic Markets. Barack Obama is currently on top.
Here's how it works: Investors can put up to $500 in an online account to buy shares in a candidate's fortunes that would pay $1 per share of the candidate succeeds, but nothing at all if the candidate fails. Likelier outcomes gain value, while unlikely outcomes dwindle in value. The idea is that the marketplace distills the "wisdom of crowds" to come up with a collective assessment that is better than any single person's opinion.
Is this starting to sound like (gasp!) gambling? Well, that's precisely the problem.
This month, the Commodity Futures Trading Commission issued a request for public comment on the future of prediction markets - or, to use the CFTC's term, "event contracts." The commission wants to know whether it should be regulating these markets, and if so, what rules should be put in place. Comments to email@example.com must be received by July 7. (Click here to read the comments submitted so far.)
On the other side of the table, researchers also want to know where they stand. Back in 1992, the CFTC issued a "no-action letter" to the Iowa Electronic Markets, affirming that Iowa's researchers could go ahead with the market as long as they followed the specified conditions.
To refresh your memory, 1992 was the last time there was a Bush in the White House and a Clinton running to replace him. Since then, no other prediction markets have been given a similar go-ahead. That's why most other real-money markets are based abroad (for example, InTrade in Ireland).
In this week's issue of the journal Science, 22 researchers - including two of the Iowa market's co-founders - call on the CFTC and Congress to clear up the uncertainty surrounding prediction markets:
"These markets could assist private firms and public institutions in managing economic risks, such as declines in consumer demand, and social risks, such as flu outbreaks and environmental disasters, more efficiently.
"Unfortunately, however, current federal and state laws limiting gambling create significant barriers to the establishment of vibrant, liquid prediction markets in the United States. We believe that regulators should lower these barriers by creating a legal safe harbor for specified types of small-stakes markets, stimulating innovation in both their design and their use."
Such assurances could take the form of no-action letters like the one issued to the Iowa researchers, but the researchers also urge the CFTC to consider issuing formal rules or guidance on prediction markets.
The Science authors aren't looking for carte blanche: They suggest that the safe-harbor treatment be extended to not-for-profit research institutions and government agencies engaged in research - as well as to "private businesses and not-for-profits that are not primarily engaged in research, which would only be allowed to operate internal prediction markets with their employees or contractors."
Among the researchers' other recommendations:
"Because Congress did not intend the CFTC to regulate gambling, it is important to design new regulations so that socially valuable prediction markets easily qualify for the safe harbor but gambling markets do not," the researchers write.
In its request for comment, the CFTC signaled that it's also concerned about the distinction between prediction futures and plain old gambling. The document spends a fair amount of ink discussing contracts that involve "commercial risk management" (such as the price of corn), as opposed to "environmental measures" (such as rainfall) and "general measures" (such as celebrity marriages).
The commission also raises the bugaboos of the prediction game: Should some event contracts be banned, such as investing in the winner of the next presidential election or the timing of the next terrorist attack? The latter idea actually came under consideration in 2003, but lawmakers quickly squelched it. You can just imagine how a senator would feel if legislation on prediction markets came up while her own stock was in a tailspin, or while someone was bidding up the market for terror attacks.
For another perspective on this, I turned to I. Nelson Rose, an expert on gambling laws at Whittier Law School (and co-author of the book "Internet Gaming Law" with Marty Owens):
"I am a big fan of predictive markets. In fact, I have an active account with the biggest, InTrade.com. ...
"The idea about getting a federal carve-out is a good one. In the late 19th and early 20th centuries, many of the states declared trading in commodity futures as being illegal gambling. So Congress passed laws expressly stating that trading stocks and commodities on a listed U.S. exchange was not gambling. It had to do the same when stock index futures were invented.
"The real problem today for sites like Intrade today is not state or federal anti-gambling law as much as they are markets for securities, which fall under federal and state securities laws. Of course, as long as they are overseas, no one particularly cares."
As much as researchers and regulators might wish it, the line between playing the market and playing the odds may not always be clear-cut. Here's an excerpt from Rose and Owens' book, quoting a participant in the Iowa Electronic Markets who made $1,300 by "investing" in Democrat Al Gore during the 2000 campaign.
"So I'm a Yellow Dog Democrat and I'm pretty sure Gore is going to capture the popular vote. On November 1, 2000, I bought about 1,900 futures on Gore at 34 cents, which cost a little over $600. On the 10th of November when the popular count was certified, these contracts were worth a buck apiece. I had earlier found the market in 1996 when the Republican Convention came here to San Diego. I made some phenomenal money 'betting' Buchanan on an uptick after the Arizona primary, and a good bunch more selling 'derivatives' of this market (i.e., campaign buttons with a little stamp on them indicating they could be redeemed for a multiple of their purchase price if the candidate won the nomination) backed by an equal number of contracts on the market.
"I'm a little surprised the Nevada books haven't offered action like this. It seems that when people put their money where their mouth is politically and actually win something tangible when their candidate comes through it's a good thing."
Is it a good thing? Or does it sound a little too unseemly when you're talking about the highest office in the land? Either way, place your bets ... er, your comments ... in the box below.
The researchers behind this week's Science article are Kenneth J. Arrow, Paul Milgrom and Erik Snowberg of Stanford University; Robert Forsythe of the University of South Florida; Michael Gorham of the Illinois Institute of Technology; Robert Hahn of the American Enterprise Institute; Robin Hanson of George Mason University; John O. Ledyard of the California Institute of Technology; Saul Levmore and Cass R. Sunstein of the University of Chicago Law School; Robert Litan of the Kauffman Foundation; Forrest D. Nelson and George R. Neumann of the University of Iowa; Marco Ottaviani of Northwestern University; Thomas C. Schelling of the University of Maryland at College Park; Robert J. Shiller and Paul C. Tetlock of Yale University; Vernon L. Smith, Philip E. Tetlock and Hal R. Varian of the University of California at Berkeley; Justin Wolfers of the University of Pennsylvania; and Eric Zitzewitz of Dartmouth College.