Science

It's Hard to Make Predictions, Especially About the Future

Why dart-throwing chimps are better than the experts

|


The price of oil will soar to $200 a barrel. A bioterror attack will occur before 2013. Rising food prices will spark riots in Britain. The Arctic Ocean will be ice-free by 2015. Home prices will not recover this year. But who cares about any of those predictions? The world will end in 2012.

The media abound with confident predictions. Everywhere we turn, an expert is sounding the alarm bells on some future trend. Should we pay much attention? No, says journalist Dan Gardner in his wonderfully perspicacious new book, Future Babble: Why Expert Predictions Are Next to Worthless, and You Can Do Better (McClelland & Stewart).

Gardner acknowledges his debt to the Berkeley psychologist Philip Tetlock, who in 1985 set up a 20-year experiment involving nearly 300 experts in politics. Tetlock solicited thousands of predictions about the fates of scores of countries and later checked how accurate they were. He concluded that most of his experts would have been beaten by "a dart-throwing chimpanzee." Tetlock found that the experts wearing rose-tinted glasses "assigned probabilities of 65 percent to rosy scenarios that materialized only 15 percent of the time." Doomsters did even worse: "They assigned probabilities of 70 percent to bleak scenarios that materialized only 12 percent of the time." 

In this excellent book, Gardner romps through the last 40 years of failed predictions on economics, energy, environment, politics, and much more. Remember back in 1990, when Japan was going to rule the world? The MIT economist Lester Thurow declared, "If one looks at the last 20 years, Japan would have to be considered the betting favorite to win the economic honors of owning the 21st century." Thurow was far from alone. In 1992 George Friedman, now CEO of the geopolitical consultancy Stratfor, predicted The Coming War With Japan. (In case you are hungry for more predictive insights from Friedman, he has recently published The Next 100 Years: A Forecast for the 21st Century.)

As oil prices ascend once again, many naturally predict that the end of this energy source is nigh. Back in 1980, Gardner reminds us, The New York Times confidently declared: "There should be no such thing as optimism about energy for the foreseeable future. What is certain is that the price of oil will go up and up, at home as well as abroad." By 1986 oil prices had fallen to around $10 a barrel. On the general accuracy of oil price predictions, Gardner cites U.S. Foreign Service Officer James Akins, who said: "Oil experts, economists, and government officials who have attempted in recent years to predict the future demand and prices of oil have had only marginally better success than those who foretell the advent of earthquakes or the second coming of the Messiah." 

Akins' observation is as acute now as it was when he said it back in 1973. In 2008 analysts at the investment bank Goldman Sachs warned that oil prices could surge beyond $200 per barrel in as little as six months. Six months later, the price of petroleum had fallen to $34 per barrel. 

When it comes to prophets, Gardner prefers foxes to hedgehogs. This distinction was made famous by the political philosopher Isaiah Berlin, who was adapting an observation by the ancient Greek poet Archilochus: "The fox knows many things, but the hedgehog knows one big thing." Foxes are intellectual omnivores obtaining disparate information where they can. Hedgehogs, by contrast, fit all information into one grand scheme that explains the operation of the world. "Hedgehogs are big-idea thinkers in love with grand theories: libertarianism, Marxism, environmentalism, etc.," writes Tetlock. "Their self-confidence can be infectious."

As Gardner shows, foxes are a bit better than hedgehogs at predicting the future. The environmentalist Paul Ehrlich, for example, is a perfect hedgehog. For more than four decades, he has maintained that everything important about the world can be explained by population trends. In his 1968 book The Population Bomb, Ehrlich notoriously declared: "The battle to feed all of humanity is over. In the 1970s, the world will undergo famines in spite of any crash programs embarked upon now. At this late date nothing can prevent a substantial increase in the world death rate."

The famines didn't happen. The world death rate was 13 per 1,000 people when Ehrlich wrote his book; it has fallen in each decade since, and it is now nine per 1,000. "In two lengthy interviews," Gardner writes, "Ehrlich admitted to making not a single major error in the popular works he published in the late 1960s and early 1970s." Yet it is not far from the truth to say that Ehrlich has never been right about anything he has predicted.

False prophets are almost never punished. Pundits like George Friedman continue to be quoted by journalists, invited to appear on TV talk shows, and hired by corporate executives eager to see what the tea leaves say about the future. When caught in an error, they emit squid-ink clouds of obfuscation, harrumphing that they got the time frame wrong, it almost happened, or an unpredictable exogenous shock derailed the forecast. Gardner reveals the recipe for achieving success as a pundit: "Be simple, clear, and confident. Be extreme. Be a good storyteller. Think and talk like a hedgehog."

In addition to making fun of the failures of the prognosticating class, Gardner cites research in cognitive psychology and behavioral economics to explain why so many of us keep falling for false prophecies: Humans hate uncertainty. "Whether sunny or bleak, convictions about the future satisfy the hunger for certainty," he writes. "We want to believe. And so we do." 

I am tempted to adopt English soccer player Paul Gascoigne's pledge: "I never make predictions, and I never will." 

Ronald Bailey (rbailey@reason.com) is reason's science editor.