Whither forecasting?

PUBLISHED : Friday, 28 November, 2008, 12:00am
UPDATED : Friday, 28 November, 2008, 12:00am

I am no more able to correctly forecast the outcome of a particular event than the next person. If you want my prediction on, say, how the political crisis in Thailand is going to pan out, you would do better drawing up a list of the probable outcomes and getting together a team of dart-throwing monkeys. I know this because, time and again, I have been asked for an opinion, given it and, more often than I care to remember, been proved wrong. For this reason, throughout most of the US presidential campaign, I plumped one day for Barack Obama and the next for John McCain: I figured that, when the results were announced, I would at least retain a modicum of respect in the eyes of half the people who had questioned me.

Please don't see my confession as a reason to write me off in the soothsaying stakes. Research has shown that I am, in fact, no worse a forecaster than the most learned analysts. People who keep abreast of current events have as good a chance of guessing an outcome as the experts. In fact, the more predictions an expert makes, the more likely it is that he or she will be wrong.

This is detailed in American academic Philip Tetlock's book, Expert Political Judgment: How Good Is It? How Can We Know? The political science professor at the University of California, Berkeley, got together 284 people who made their living 'commenting or offering advice on political and economic trends' and asked them for their projected outcomes of a range of events. A statistical quantifier was introduced by putting most of the questions in a 'three possible outcomes' format: that the status quo would be maintained; there would be more of something; or less. The experts were measured in terms of guessing probabilities and how accurate they were at predicting specific outcomes.

I felt bad about my skills of prediction until I read Professor Tetlock's results. On the first scale, the analysts did worse than if they had randomly chosen one of the three outcomes. They were also found to be no more accurate in guessing how an event would end than non-specialists. The upshot is that there is no reason to suppose that I - or South China Morning Post readers - know any less about Thai politics than those who specialise in the country.

If this is the case, why then do we so avidly read the slew of crystal-ball gazing that is churned out by academia, think-tanks, government departments, non-governmental groups, business research units and the like? How can the US National Intelligence Council be better versed than this humble correspondent in forecasting what the world will look like in 17 years, as the organisation has done with its report, 'Global Trends 2025: A Transformed World?' I wager that, for all the research, analysis and co-operation among agencies, I could make as good a guess. Even a monkey with a dart and an outcomes list could do as well.

All such studies generally reach the same conclusions. They are that: rising powers China, India, Brazil and Russia will set the future global agenda; the flow of capital from west to east will continue to grow; the rising world population will put pressure on resources like land, water and food; we need to do something about religious extremism, nuclear proliferation and disease if the world is to thrive; and we are living in fast-changing times and should not expect the situation to vary.

We've heard these predictions time and again: anyone could have made them with reasonable accuracy. Perhaps the analysis industry should take a hard look at itself and rethink its purpose.

In troubled times, I turn to the King, Elvis Presley, for sage advice. This time, it can be found in his song, A Little Less Conversation. The lyrics include the lines:

A little less conversation, a little more action please

All this aggravation ain't satisfactioning me

A little more bite and a little less bark

A little less fight and a little more spark.

So, perhaps instead of coming up with predictions that any of us could make, the hordes of professional experts the world over should start telling us how to solve the problems ahead.

Peter Kammerer is the Post's foreign editor