Print this article |
Close this window
Every crowd has a golden lining: bet on it
June 23, 2009 - 7:08PM
Want to know what the future looks like? Follow the crowd and the money. It's an idea that might explain the growing
interest by economists, business and policy pundits in predictive markets and their forecasts on everything from election
results to business ventures, writes Leon Gettler.
These forums - also known as internal, information, opinion or event markets - work by aggregating otherwise decentralised
data from a collective. The group gets it right more often than the smartest individual.
At the end of last year, global research and advisory firm Gartner forecast that a combination of predictive markets, wikis -
special websites on which anyone can post material - and free open-source software would drive a 10 per cent rise in corporate
productivity in the next 10 years.
Gartner's sweeping claim ignores that collaborative work has been around for decades. The Total Quality Movement in the
1960s and '70s in Japan, for example, encouraged assembly line workers to innovate and improve routine production processes.
What's different now is that technology makes decentralisation easier. That suggests more organisations will be adopting
decentralised processes and market-generated forecast systems.
A working paper released last year by US academics Justin Wolfers and Eric Zitzewitz found these systems consistently
outperformed established benchmarks. For example, the Iowa Electronics Futures Market, which predicts votes for election
candidates, had a 1.5 per cent error margin; Gallup Poll's was 2.1 per cent. Regulated by the Commodity Futures Trading
Commission, the Iowa system works as a real-money winner-takes-all market where people can bet on candidates by buying
contracts based on those candidates winning. The market for contracts ultimately picks the price and probability of the result.
Similarly, in the last federal election here, opinion polls were pointing to a close-run thing. The punters knew