Came across this interesting little paper. The title can be a little confusing since the term prediction market is simply used to convey a market that generates odds. When they say prediction market think bookmaker.
The gist of the paper is that such markets have been used for centuries for a variety of events but have been largely ignored in financial or economic forecasting in favour of professional economists. It doesn’t take much insight to realise that such prediction markets outperform economists.
From the papers conclusion.
Over the last half decade, many economists, and the public, have re-evaluated the eﬃciency of ﬁnancial markets. While these re-evaluations have not been favorable to markets, it isimportant to keep in mind the alternative (Zingales, 2010). In the case of forecasting, the alternative is often professional forecasters, polls, pundits, or a combination of the three.
As we have shown in these paper, prediction markets out-perform both professional forecasters and polls in a variety of statistical tests.We have shown that prediction markets have many of the properties expected under theeﬃcient markets hypothesis. In particular, they are diﬃcult to manipulate, lack signiﬁcant arbitrage opportunities, aggregate information quickly and in a seemingly eﬃcient manner.
Evidence of eﬃciency can be seen in the macro-derivatives markets, which out-perform professionalforecasters, or in political prediction markets, which out-perform polls.