« The Triumph of Authenticity in "Thank You for Smoking" | Main | Is the future united? »

Sunday, May 07, 2006

How to Earn Free Money on the Tradesports Presidential Prediction Market

Political junkies and ardent free marketeers probably know about prediction markets  such as tradesports and the Iowa electronic market. For those with something better to do with their lives, here's a brief explanation of the concept from wikipedia.

Also known as information markets, decision markets, idea futures, and virtual markets, prediction markets are speculative (i.e., betting) markets created for the purpose of making predictions. Assets are created whose final cash value is tied to a particular event (e.g., will the next US president be a Republican) or parameter (e.g., total sales next quarter). The current market prices can then be interpreted as predictions of the probability of the event or the expected value of the parameter.

People who buy low and sell high are rewarded for improving the market prediction, while those who buy high and sell low are punished for degrading the market prediction. Evidence so far suggests that prediction markets are at least as accurate as other institutions predicting the same events with a similar pool of participants.

Prediction markets have been lauded by books such as The Wisdom of Crowds. As an economist, I love the idea that a free market does as well or better than individual prognosticators or polls in predicting the outcome of elections. But my latest glance at the 2008 U.S. presidential market has dampened my enthusiam about electronic markets. Here's why.

Tradesports is selling futures allowing investors to bet on which party will win the 2008 election. A security that pays one dollar if the Democratic party candidate wins is currently selling for 49.1 cents, a security that pays one dollar if the Republican party candidate wins is selling for 48.5 cents, and a security that pays one dollar if the winner comes from a party other than the Republicans or Democrats is selling for 2.4 cents. If you add up the prices of the all the securities, they equal $1.00 (=.491+.485+.024).

This makes no sense. These prices suggest that, collectively, investors are willing to pay one dollar today in exchange for one dollar in November 2008. (If I bought all three securities at today's market price, I would be assured of getting a dollar after the 2008 election was finished, and I would have to pay a dollar today.)  In a normal market, this would never happen. The time value of money implies that the value of one dollar in November 2008 would be less than one dollar today. 

In a normal market, arbitrageurs would have a field day. They would sell all three contracts (Democrat, Republican, and other) at the current price, netting a dollar. The arbitrageurs would then invest the money and earn interest until November 2008, at which point they would pay one dollar to those who bought the security of the party that ultimately won. The arbitrageurs would pocket the interest earned between today and November 2008. This is a way to make money with no risk. No matter what happens, the arbitrageur only must pay out a dollar in two and half years.

Free money shouldn't last long. Arbitrageurs (or tradesports itself) should be offering securities by the bucketload, driving down the prices of all three securities until their combined value falls well below one. The fact that this isn't happening suggests that the prediction market is not quite as logical and efficient as claimed.

So what is going on. There is probably a norm in the markets that the value of all the securities should add up to one. But this is an incomplete answer. Even if there is such a norm, arbitrageurs should try to take advantage of the norm as described above. So perhaps arbitrageurs cannot enter the market? For example, if only tradesports can issue the securities, then arbitrageurs cannot play the strategy described above. But this is still an incomplete answer. Why doesn't tradesports itself flood the market with new securities and effectively get interest free loans? Perhaps tradesports is concerned that people would be turned off by a market in which the value of all possible outcomes was less than one, and therefore tradesports restricts supply.

Perhaps, but something is amiss here. Why are the investors in prediction market securities willing to give interest free loans? I would love to here other explanations, but for the moment I am concerned that prediction markets (or at least this particular well known prediction market) are not all they are cracked up to be. 

Posted by Yair Listokin on May 7, 2006 at 09:36 PM in Law and Politics | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference How to Earn Free Money on the Tradesports Presidential Prediction Market:


Dickinson's observation is enough to refute the post. There's also the bid-ask spread, and most markets of this sort have a transactions fee. So there's lots of room for friction inefficiency preventing arbitrage.

Posted by: Ted | May 9, 2006 12:51:17 AM

McCain loses the GOP nomination, runs as an independent, and wins--not obviously less than 2.4% likely.

Posted by: Chris | May 8, 2006 5:17:19 PM

Is Fir a spammer?

Posted by: Eh Nonymous | May 8, 2006 12:11:16 PM

you bring up an interesting question. just a thought (and i may be misunderstanding the finance computations here): tradesports charges a $0.04 trading fee for price takers, so if you were to short the three contracts at the current bid price, you would not receive a full dollar. are there other transaction costs that could explain it?

Posted by: siona | May 8, 2006 1:52:12 AM

Tradesports requires you to have enough money in your account to cover the worst possible loss scenario, which is "frozen" when you sell a contract. So in your hypothetical, the $1 would be frozen and you couldn't invest it. If you deposit over $20,000 you can earn 3% interest, although there isn't enough volume in that market alone for your freeroll to make up for the interest you'd lose by keeping the $20,000 in such a low-yield position. Of course, this doesn't change the fact that prices should be much lower. As a rule, I think the low-volume event markets on Tradesports tend to be overpriced regardless of duration.

As a sidenote, Tradesports contracts are for $10, not $1.

Posted by: John Dickinson | May 8, 2006 1:41:45 AM

Any market where people are paying 2.4 cents for a neither party contract has a far more fundamental disconnect with reality.

Posted by: Dylan | May 8, 2006 12:16:09 AM

The comments to this entry are closed.