Surowiecki claims that if you ask a large enough sample of diverse, independent people to make a prediction or estimate a probability, and then average those estimates, the errors each individual makes in coming up with an answer will cancel themselves out. Each persons guess has two components: information and error (or signal and noise). The process of averaging tends to cancel out the errors, leaving the information.
On the other hand, small groups of experts tend not to have a diversity of opinion and nor have they necessarily come to their opinions independently. Experts are often vulnerable to “groupthink”. This was evident in the period before the global financial crisis – which 99% of economists failed to predict. Thus, it is hard for business and government planners to manage risks effectively.
We have found that the concept of the Wisdom of the Masses is rather more complex that suggested by Surowiecki. In addition to the signal and noise components, most individuals are prone to bias of various sorts. These biases do not cancel out when taking an average, so to forecast with the wisdom of the masses, we have to exclude biased opinions. The Wisest of the Masses are a small segment of the total population whose predictions are judged to contain little bias.
The Wisest of the Masses shows the potential to be able to forecast GDP more accurately than panels of economists.
The opinions of the masses are important in two other ways, not mentioned by Surowiecki. This includes using consumer expectations to forecast.