I've been doing a lot of reading lately; Malcolm Gladwell, Steven Levitt, Dan Gilbert among others. They all have fascinating things to say. However the more I read, the more I'm left with the feeling that while the stories they have to tell are intriguing, they are grossly simplified.
I've just picked up Daniel Gardner's "The Science of Fear" and am reading the prologue. In it he talks about how the September 11 attacks caused American's behaviour to change.
It was an unreal, frightening time, and it was predictable that people would flee the airports. Perhaps surprisingly, though, they didn't start digging backyard bomb shelters. Instead, most went to work and carried on living. They just didn't fly. They drove instead.
Politicians worried what the mass exodus of Americans from planes to cars would do to the airline industry, so a bailout was put together. But no one talked about the surge in car travel. Why would they? It was trivia. There were deadly threats to worry about.
But what no politician mentioned is that air travel is safer then driving. Dramatically safer — so much so that the most dangerous part of a typical commercial flight is the drive to the airport.
The safety gap is so large, in fact, that planes would still be safer than cars even if the threat of terrorism were unimaginably worse then it actually is: An American professor calculated that even if terrorists were hijacking and crashing one passenger jet a week in the United states, a person who took one flight a month for a year would only have a 1-in-135,000 chance of being killed in a hijacking — a trivial risk compared to the annual 1-in-6,000 odds of being killed in a car crash.
…
It turned out that the shift from planes to cars in America laster one year. Then traffic patterns went back to normal. Gigerenzer also found that, exactly as expected, fatalities on American roads soared after September 2001 and settled back to normal levels in September 2002. With these data, Gigerenzer was able to calculate the number of Americans killed in car crashes as a direct result of the switch from planes to cars.
It was 1,595. That is more then one-half the total death toll of history's worst terrorist atrocity.
…
And yet almost nobody noticed but the families of the dead. And not even the families really understood what had happened. They thought — they still think — that they lost husbands, wives, fathers, mothers, and children to the routine traffic accidents we accept as the regrettable cost of living in the modern world.
They didn't. It was fear that stole their loved ones.
It's easy to look at that story and those numbers and come to the conclusion that humans are bad at estimating risk and let fear rule their decisions. But it's not that simple.
On a population basis it's a compelling and important piece of information. On an individual basis it misses out on some critically important pieces of information which people absolutely should factor into their decision making process:
The September 11 attacks meant that something "new" happened, it's only in hindsight that we can talk about it as an isolated event. At the time when everybody was making decisions about how to travel, nobody knew if there were a long line of attacks planned or how effective American security procedures would be at thwarting them. In this context it makes complete sense for people to take a step back and wait to see what happens next.
When you get onto a plane you are playing a trust game. From the moment a plane takes off to the moment it lands, there is almost nothing you can do to influence your safety. Trust games are fine (even fun!) when you are comfortable with the risks, when there are new risks that you don't understand ... not so so much.
Driving, while inherently more dangerous then flying, offers a degree of involvement and control that simply doesn't exist on a flight. Have your kids in the car, drive slowly and carefully. Don't like the traffic, pull over. Tired, pull over. Something crazy happening on the news, get a motel room and hide. Volcano erupting, drive for the hills.
To me the change in driving habits represents an entirely rational decision making process. In times of risk revert back to things which you understand and can control. Billing this behaviour as "driven by fear" seems simplistic in the extreme.
PS. Don't get me wrong. I use the the difference in statistical risk between flying and driving in conversation all the time, it's a great way to demonstrate the difference between perceived risk and actual risk. I also get that the above information could have been used by politicians to save lives by shaping policy decisions. I just don't buy fear as the only explanation.