Prediction Markets: a Source for Quality Information?

In the search for meaningful and reliable information to support quality decision making, prediction markets have an intuitive appeal. Employ the collective wisdom of a crowd to assess the probability of a future event.

Setting up a prediction market is conceptually simple. For an uncertain but well-defined future event (eg, “Barack Obama is re-elected President of the United States in 2012”), with a specific expiry date, create a security (financial instrument) that pays out 100 of your favourite monetary unit— mine is the pound sterling—if the event occurs on or before the expiry date. If the event does not occur, the security pays out nothing.

The price someone is prepared to pay for such a security should reflect the probability the buyer assesses for the event. Get enough people together to trade the security and the market pricing mechanism will likely provide a pretty good assessment that we can use in decision making. (Let’s ignore for a moment the finesse required to adjust for time preference. A promise of £100 in November 2012 is worth less than £100 today.)

Studies have shown the accuracy of prediction markets is assessing the chances of political outcomes. And various companies are using internal prediction markets to forecast sales and the success/failure of R&D projects.

What prompted this blog was spotting, just a few days after writing about Tony Blair’s decison making, a graph of the market on “Will Weapons of Mass Destruction be Discovered in Iraq?” On the eve of the war, the implied probability assessment was about 70%.

Source: Prediction Markets by Justin Wolfers & Eric Zitzewitz, 2004

I am fascinated by the potential offered by prediction markets to inform decision makers. Now I need a client who agrees and wants to explore this exciting area further. Anyone?

See also:

• Prediction Markets by Justin Wolfers & Eric Zitzewitz. Their 2004 paper includes the Iraq WMD prediction market graph.
• Betting to Improve the Odds by Steve Lohr of the NY Times. An interesting article about major companies using prediction markets to explore the potential of new ideas.
• The Wisdom of Crowds by James Surowiecki explores the fact that, under the right circumstances, groups are remarkably intelligent, and are often smarter than the smartest people in them. Prediction markets are one way of harnessing that intelligence.
• InTrade, formerly TradeSports, is regulated in Ireland and offers the public traded contracts on a wide range of events, including the discovery of WMDs in Iraq.
• CrowdCast and MIT have set up Pharmer’s Market to develop forecasts of the chances of success of various breast cancer drugs in Phase II/III development. Read CrowdCast’s 9-Oct-09  press release.
• NewsFutures helps companies set up prediction markets with their employees, customers and suppliers. The website features an excellent collection of links to articles, papers and books. It also hosts public markets for predicting baseball and American football results, but frankly, who gives a s**t about minority sports such as these?