On economists as narrow-minded hedgehogs; the perils of futurology

In December 2007, when BusinessWeek magazine published its annual chart of economic forecasts for the following 12 months, every one of the 54 economists surveyed predicted that the US economy would not “sink into a recession”. It would, they all declared, be a solid but unspectacular year. They weren’t alone in failing to see what was coming. In July 2009 a group of eminent British economists and constitutional experts wrote to the Queen bemoaning “a failure of the collective imagination” that had led none of them to predict the credit crunch.

Much of the fun of reading Dan Gardner’s oddly cheering book stems from the attempts of pundits to wriggle out of their duff predictions after the event. Gardner, a Canadian journalist, wants to know why expert predictions are so regularly so wide of the mark, and why there’s rarely a reckoning with those who have got things wrong.

Future Babble takes its cue from the work of Philip Tetlock, an American psychologist who grew so annoyed with cocksure punditry that he devised an unusual experiment. In 1984 Tetlock selected 284 experts – academics, economists and think-tankers – who made their living hawking advice on political and economic trends and asked them for testable predictions about how the future might pan out. He went back to check 20 years later, and the results were not impressive. Most experts, Tetlock showed, are no better than “dart-throwing chimpanzees” at the prognostication game. Worse, when challenged about their mistakes many of them failed to own up – some tried to fudge, while others claimed to have been right all along.

Part of the problem, Gardner explains, lies with a lack of accountability within the profession itself. Make a bold prediction and the journalists and TV cameras come running; no one remembers when it fails to come about. Following Isaiah Berlin’s celebrated distinction, Gardner divides pundits and forecasters into two kinds of beast – the fox who knows many different things, and the hedgehog who knows one big thing. Hedgehogs, he says, have a narrow range of expertise and tend to arrive at bold, bullish predictions. The forecasts of hedgehogs are simpler and more entertaining, so they soak up all the media attention. But they are much more likely to be wrong than foxes. With a wider range of data and disciplines to scavenge from, foxes tend to be more careful with their predictions, and to fare better as a result.

It isn’t only economists who suffer regular bouts of prediction failure. Who, hand on heart, can say that they predicted the timing of the Arab risings? Demography is a staple of futurological punditry, but even here forecasting is not as easy as it looks – hardly anyone, Gardner points out, reckoned on the baby boom that followed the second world war. The modern environmental movement has been stymied by the regularity with which its prophecies of imminent catastrophe fail to come true. Paul Ehrlich’s 1968 book Population Bomb, for example, argued that the earth’s population was growing more quickly than our ability to feed it – and painted a terrifying picture of the era of mass starvation to come in the following decade.

Reading this hugely enjoyable litany of failed predictions, it’s easy to emerge with the impression that predicting the future is a fool’s errand. Some things, however, are eminently predictable, as life insurers and those who control our supermarket loyalty cards well know. Good advice for the budding forecaster is not to try to fix things that are necessarily in flux. Avoid talk of time bombs, and steer clear of predicting the end of anything – wars, resources and certainly not history. And try not to peer too far into the future, or you’re very likely to fall flat on your face.

Dan Gardner’s Future Babble is published by Virgin Books. This review originally appeared in The Financial Times, 14-15 May. James Harkin is director of Flockwatching, and author of ‘Niche: Why the Market No Longer Favours the Mainstream’ (Little, Brown)

Leave a Reply

Your email address will not be published.