It’s becoming almost passée in the econ blogosphere to show graphs from prediction markets regarding the coming US presidential election (only 10 months to go!!). Anyway, here’s my contribution with the latest action on the Intrade markets for the Democratic party nomination, following Hillary’s win in New Hampshire:

Now the question is, was Obama an inefficient market bubble, or an efficient market responding to the best information that was available at the time, plus a sudden unexpected event?

I noticed another funny thing with the Intrade data. The prices are probabilities, and if you add up the prices for Clinton, Obama, Edwards and Gore to win the Democratic nomination, you sometimes get more than 100, which should be impossible:

Since a total over 100 is logically inconsistent, doesn’t this represent some sort of arbitrage opportunity or something? However, the excursions above 100 are brief, so maybe the market is pretty efficient.

Update: Felix Salmon identifies some real arbitrage opportunities on Intrade. Maybe it’s not so efficient after all.

I was wondering about the extent to which blog chatter correlates with information from other sources such as prediction markets. This is pretty difficult to test, because the things that people write on blogs can be hard to decode and categorise. But I think I found one reasonable and straightforward test case: use of the word ‘recession’ in blogs versus the probability of a recession from a prediction market. The theory is that when a recession is more likely, people should blog about it more, and the level of recession chatter should be positively correlated with the probability of a recession predicted by a prediction market.

So I got the daily counts of the number of times ‘recession’ was used in blogs, according to Technorati, and the probability of a US recession in 2008 according to Intrade. The overlapping data for both series covered about the past 90 days. Of course, people could blog about recessions in any country, not just the US, and people could use the word ‘recession’ when they’re saying something like ‘a recession is unlikely in 2008′, so I expect the relationship to be far from perfect. Here’s the basic data that I got, with the count of the word ‘recession’ on the left, and the Intrade probability on the right:

Here’s a scatter plot of the two data series together, with a fitted logarithmic curve (it fits slightly better than a linear relationship):

For those econometricians reading this who may be thinking spurious regression, I did a Phillips-Ouliaris cointegration test on the fitted model above and the p-value was 0.079, so it’s borderline whether this is spurious or not, but I’m willing to believe there’s a meaningful relationship.

The creator of the Financial Next website asked me to take a look at his site. It’s a website that facilitates ‘prediction markets‘ with play money (ie not real money). This is not a new idea, for example the Iowa Electronic Markets have apparently been around since 1988. The basic idea is to create a virtual online market for a well-defined future event, for example ‘Hillary Clinton wins the 2008 US presidential election’. For a yes/no event like that, if you own a share in that idea, you get some payoff, say $100, if it turns out to be true, and nothing if it’s not true. How likely you think the event is will determine what you should pay for the share. If you think Hillary has an 80% chance of winning, then you should be willing to pay up to $80 for a share in that idea.

Aggregated over many people, the market price in such an idea converges on the overall ‘consensus’ opinion of all the traders in the market. Basically, markets are tools for extracting information. In this case we’re extracting people’s beliefs about the probability of some unknown future event. We could do the same thing by just taking a survey of people’s opinions. However, there’s some reasons why markets might generate more accurate results than surveys. First, if money is at stake, people might take the question more seriously and try to come up with an informed opinion about the issue in question. Second, it makes clear the distinction between “Do you want Hillary to win?” and “Do you think Hillary will win?”. There are probably other reasons why markets might outperform surveys, but those are the two that come to mind for me.

On the downside, it’s probably more expensive to run prediction markets than conducting surveys, and people have to be willing to spend more effort participating in them. You can answer a survey question in one minute, but a prediction market probably requires more intensive effort over a period of time. Thus it might be easier to gather the opinions of more people with surveys, compared to what could be achieved with prediction markets. I’m not sure whether surveys or prediction markets would win in a cost-benefit analysis of the tradeoff between cost of running the process and accuracy of the results.

In any case, as an economist I like the idea. The Financial Next site implements this pretty well and makes it easy for users to participate in markets, and even create their own markets. Since it’s just for play money, there’s no real harm done if people set up strange markets on the site. On the other hand, I wonder how seriously people take it when only play money is at stake. As the site says, the main motivation for participating in the site is fun and beating other people. So it’s a kind of game.

I think this is another interesting idea — using games to motivate people to participate in economic experiments. These days experimental economics is popular, and economists try to set up artificial situations with subjects in a lab to test various economic theories. One criticism of these is that the subjects are not really motivated enough and won’t behave as they would in a real situation. I think well-designed games (here I mean not the game theory type of game but the ‘fun’ type of game) may be able to help overcome this problem. If people enjoy playing a game, and have a stake in it, they may take it seriously, and if the game it set up well maybe we can test economic theories. In other words, give people a fun game as a carrot, but manipulate the environment in some way to run some subtle experiments. One online game, EVE, recently hired a real economist, and it seems like he’s having some of these same thoughts.

In terms of the Financial Next site, its main problem at the moment seems to be a lack of traders. Since it’s only for play money, people will only be motivated to join if the site provides them with some other kind of benefits aside from making money. The competitive/play motivation is fine, but that works best when there’s a large population of players. If I’m the best out of 10 players that’s not so impressive, but best out of 100,000 is something to be proud of. So Financial Next is another example of a network that will become more valuable to its members as it gets bigger. I’m not sure exactly how they can attract more users and reach some sort of ‘critical mass’, but I’d suggest emphasising the competitive aspect more and trying to make trading more ‘fun’ somehow.

The other problem with a lack of traders is that the market becomes quite ‘thin’. This means that its predictions won’t be so good, because they don’t aggregate the opinions of very many people. In addition, traders’ behaviour probably becomes somewhat strategic. In a small market with few traders, each of them has quite a lot of influence over the market price. This can affect players’ strategies, and may lead them not to trade in such a way that best reveals their estimates of the probability of the event that we’re interested in. In contrast, in a thicker market with many traders, each will assume that they can’t really affect the price, and will be more likely to trade according to their true estimate of the probability.