What Is Not Seen

An econ log on financial markets and the global economy.

Posts Tagged ‘Politics

Predicting the 08-presidential election

without comments

With the election day getting closer, predictions about the outcome are getting plenty. The latest pollster numbers indicate that more than 80% of Americans think the country is on the wrong track, and only 28% believe that the president is doing a good job. From numbers like these, one can easily draw conclusion that the election already is a done deal. But how good a prediction are they really? As every other survey, there is always a margin of error: people may answer dishonest, or may have a vested interest on one side, etc. Also, questions like: If the election were held today, who would you vote for? do not account for the fact that the elections are held in the future. Some thing else is clearly needed. 

When it comes to the business of forecasting prices, future markets surve as a platform where traders agree to buy or sell merchandise at some future date. Reasonably such interaction should result in better predictions, than say pollsters. As Hayek pointed out, market prices have a dual role, except for allocating resources to their most productive use, they also contain information of their value. When people have a vested financial interest at stake, the above mentioned problems with polls could easily be avoided with futures.

This brings me to IEM or the Iowa Electronic Markets. Like future markets such as CME (Chicago Mercantile Exchange), IEM provides a futures market for political election. By looking at future prices for the current presidential election, we get a better picture of the probable outcome. By watching the price paid for a democratic versus a republican victory-future, we get the percentage probabilities for both parties geting the majority of votes, as can be seen in the picture belove:

We can see that the Democrats have been in favor, pretty much all the time since June 14th 2006, when the history started. But looking at the graph we see that the republicans gained support early on this year. 
At the moment, the likelihood for a Democratic victory are 62.8 percent probability versus 32.5 for a republican victory. 

The election is held 4th of November later this year. Who will win is by no way certain. Lots of things can still happen. Don’t just listen to the pollsters, and do remember to check the IEM futures market in the upcoming months!

Written by Daniel Halvarsson

July 4, 2008 at 5:06 pm

Posted in Economic Analysis

Tagged with