There has been much talk over the past year or so of the public flagrantly ignoring the advice of the experts on public policy such as Brexit. What is surprising is that those supporting the experts seem to have so little knowledge of how well experts perform on matters of public policy.
Philip Tetlock summarised research on the accuracy of expert predictions in his book Expert Political Judgement. How good is it, how can we know?, the conclusion was that experts performed overall worse than chance on predictions of public policy. The research shows that trusting the experts to predict outcomes “…is worse than putting dart-throwing chimps in charge of public policy.” (Zeljka Buturovic).
The best predictors seem to be crude algorithms that extrapolate from the present to the future – ie: things will be continuing roughly in the direction they have already taken. Extrapolations from current trends were almost twice as accurate as the experts. The experts actually performed worse than “dilettantes” who are people who are well read and qualified but not actual professional experts in a field. The worst performers in the studies were ordinary undergraduate students.
We should all ask how anyone could consider themselves an “expert” who does not focus on current trends. In economics and the social sciences the dismissal of the continuation of simple trends seems to be a widespread fault.
This failure of the experts should not surprise us. Public policy is complex. What should surprise us is that the UK media value “expert predictions” above current trends, especially when those trends contradict expert opinion and are widely available. Given that the trade deficit with the EU is already costing 4% of UK GDP (an £80bn real reduction in GDP now) the media might be cautious about expert predictions that suggest Brexit will actually make this situation worse.
Raising the status of “predictions” above that of actual trends would seem to be simple political bias on the part of the media.
Social media is a factor that has altered the media response to public policy prediction. Media pundits garner the public mood from Twitter, Facebook etc. However, they fail to take account of the fact that these media are dominated by the young and inexperienced (undergraduates) and also fail to notice that much social media is the public face of employees. Many managers now ask employees to be “Friends” on Facebook and many people allow co-workers as “Friends” which immediately distorts the user’s pages into confirmations of corporate attitudes. Corporate attitudes are often anti-Brexit because large multinational companies are heavily invested in the EU.
Editors and journalists in the broadcast media are surrounded by expert predictions, student social media postings and corporate propaganda and have opted to favour the expert predictions that confirm the bias in their media environment over the current trends. The bias towards expert prediction is so marked in the case of the Brexit debate that broadcasters suppress all mention of current trends.
There is an excellent account of the accuracy of Treasury forecasts for Brexit at this link.
This post was originally published by the author on his personal blog: http://pol-check.blogspot.com/2018/02/should-media-have-rallied-round-experts.html