Try Your Hand at Economic Forecasting

Think you can do a better job at predicting the economic future than all those economists and pundits?  Here’s your chance to prove it:

Members of the public are being encouraged to take on the Bank of England by betting on the U.K.’s future inflation and unemployment rates.

Free-market think tank the Adam Smith Institute on Wednesday launched two betting markets in an attempt to use the “wisdom of crowds” to beat the Bank of England’s official forecasters. Punters can place bets on what the rate of both U.K. inflation and unemployment will be on June 1, 2015.

Sam Bowman, the research director of the Adam Smith Institute, believes the new markets will “out-predict” official Bank of England predictions.  “If these markets catch on, the government should consider outsourcing all of its forecasts to prediction markets instead of expert forecasters,” he said.


My prediction: the prediction markets will beat the experts by a little, but still not be sufficiently accurate or consistent to provide a dramatically more reliable benchmark for future inflation/unemployment rates.

Why? From what I can tell, the future inflation and unemployment rates are functionally chaotic beyond a certain level of detail. Broad trends are able to be forecast with reasonable accuracy, but the fine details have too many factors to be well predicted.

The prediction market might consistently get a tenth of a point closer (on average across a large sample of predictions) to what actually happens than the experts, but a tenth of a point (on average) is not a sufficient margin to significantly change policies.

Gwen Shapira

The Bank of England's predictions are public and the bettors can (and probably should) take them into account when placing their bets.

This means that even if the markets out-predict the official Bank of England predictions, we cannot conclude that they can replace the expert forecasters.


I may be wrong on this, but I feel like predicting inflation and unemployment requires more information than most of the people who are betting would need to make a choice. In that case, information on expected changes in inflation and unemployment would be like a public good--under supplied in the market (crowd) and left to freeriders to bet with what they hear from analysts who have the most information... In other words, it's not really the Bank of England vs betting markets--it's Bank of England vs the consensus of a handful of people with inflation and unemployment information in front of them with the popular support of informational freeriders. Any thoughts?


It would be fun but I think they got the wrong product. "Outsourcing" will not work in the case of the BoE or any similar institution, they do not need the predictions themselves as much as they do need the capacity to analyse and predict. The bulk of the job is to run internal scenarios, after all this is an institution that not only predicts but influences the state of the economy. The official forecasts are but a small. Art of the total work done.
Not to mention reliability (or even availability) of forecasts -- what happens if most of the betters decide to take August off, not place a bet while on vacation? Is a thin market better than the official forecasts?

Steve Cebalt

The Folly of Predictions: