# Air crash or urban crime—which is more likely to get you?

The risk of being killed in an air crash and the risk of being killed as the victim of an urban crime both have declined dramatically in recent years. And yet, while one is still much more likely to happen than the other, we disproportionately focus and worry about the least likely—of the two—way we could die, concludes MIT Sloan Professor Arnold Barnett.

With an expertise in applied mathematical modeling, focused on health and safety, Professor Barnett presented “Mortality Risks” in a workshop for the School’s alumni during Alumni Weekend 2011.

Barnett’s preference for measuring the safety of air travel is to calculate the death risk per randomly chosen flight—taking a particular set of flights and choosing at random from among them, what is the probability of death because of a problem with the airplane?

In the decade 2000-2009, the worldwide death risk per passenger flight was 1 in 3 million, but that is derived using an average of death risks from around the world, which could vary greatly. In the United States from 1994 to 1996 the death risk was 1 in 3.6 million. During that time, The Federal Aviation Administration set a goal of reducing that risk by 80 percent. In 2007, the risk of death was 1 in 32 million, an 89 percent reduction, showing the risk of dying in an air crash is significantly lower in the U.S. today than it was less than 20 years ago.

Barnett compares that statistic to the likelihood of growing up to be president of the United States. Figuring that on average a new president has been elected every 5.3 years since George Washington was first elected, and that currently there are 22 million births every 5.3 years in the U.S., a child’s chance of growing up to be elected president is about 1 in 22 million.

“An American kid is more likely to grow up to be president than to perish on the next flight,” Barnett says. “That’s one measure of how extraordinary a job has been done” in improving air travel safety.

From another perspective, Barnett offers, if a person flies daily, it would be 88,000 years, on average, before that person dies in an air crash.

“The accidents have been brought to the brink of extinction,” he says. Thus, dying in an air crash is not so much a risk, as an obsession.

The risk of dying as the victim of an urban crime has also declined in recent years. Barnett used 2007 data from the 50 largest cities in the United States, excepting New Orleans, because of post-Katrina data problems, and Las Vegas, because suburban data is included in Las Vegas’ information. The numbers vary greatly by race and gender with the extremes being the risk for a black male highest and the risk for an Asian female being lowest. The risk for those groups decreased since 1982 at a rate ranging from 29.1 percent for black males to 71.4 percent for Asian females. The 2007 data showed the risk for a black male to be 1 in 23 and the risk for an Asian female to be 1 in 952. The overall lifetime victimization risk in those 50 cities is 1 in 130.

Thus, the chance of being killed in an urban crime is 1 in 130 and the chance of being killed in an air crash is 1 in 32 million.

“(Homicide) continues to be a tragedy of terrible proportion,” Barnett says. “Homicide is a much larger source of risk in the U.S. than airplane flight is. But homicide is a background tragedy that goes on and on and on and people don’t know too much about it. There’s something odd that we seem to focus on low risk events and pay little attention to risks that are much higher.

For Barnett, this leads to a bigger question.

“Are we making mistakes that have massive consequences because our limited resources are being directed against phantoms?”