Mark Stewart and John Mueller are here to warn us about the security at our airports. They want to warn us that it is too good. Or, at least, there is too much of it. Their new paper is titled: “Cost-benefit analysis of airport security: Are airports too safe?” The answer, they say, is most likely yes.
Stewart is an engineer at the University of Newcastle in Australia and an expert in risk-modeling. Mueller is a political scientist at Ohio State. They have written books, papers and essays arguing that Americans greatly overestimate the risk of and impact from terrorist attacks, and have allowed that fear to distort their lives (Mueller also, charmingly, has written a much-praised book on the dancing of Fred Astaire). The point of the new paper is to do the sort of cost-benefit evaluation of airport security that gets done by engineers and insurers and policymakers in other settings. “This kind of analysis is pretty standard for other hazards,” says Mueller. “For example, if you’re planning to build some underground shelters for tornados in Alabama, the questions are exactly the same: how many lives do you save, how much does it cost, what is the likelihood of tornados?”
Stewart and Mueller calculated the cost of traditional airport security measures and measured it against the risk of an airport attack, the cost of the damage an attack would cause (in lives and property), and the efficacy of particular security measures in preventing an attack. Their finding: “Many of the assessed security measures would only begin to be cost-effective if the current rate of attack at airports in the U.S., Europe, and the Asia-Pacific increases by a factor of 10-20.”
Reading the paper, two things become clear. One is that it confines itself to security measures that are meant to protect airports themselves, not airplanes. So the TSA scanners and pat-downs and ID checks, the air marshals, all of that stuff is out of their purview. Still, there are many measures we take to protect air terminals as spaces—police patrols, blast barriers, not allowing cars to linger at the curb. And in the wake of November’s LAX attack, there is talk of adding additional security measures.
The second thing to point out is that Stewart and Mueller are looking at measures proposed to strengthen airport security, things like vehicle search checkpoints, hiring more skycaps to keep an eye on vehicles, adding curbside blast deflection and shatterproof glass, closing off the car lane closest to the terminals, deploying more bomb-sniffing dogs. The paper doesn’t attempt to do a cost-benefit analysis of all the actual security measures now in place at the world’s airports. The authors’ skepticism about those—and the implied answer to the provocative question in its subtitle—is their extrapolation.
As Stewart and Mueller readily concede, there are all sorts of assumptions that go into this sort of calculation. To account for that, they took care, they say, to give the benefit of the doubt to proponents of more airport security. In their calculation, they purposely inflate the benefit side of the comparison as much as possible, using figures at the high end of the scale for how likely attacks are, how much damage they cause, and how effective security measures would be in preventing them. Even after all that, they argue, the measures still look like a waste.
Critics of this kind of cold-blooded calculation (Mueller and Stewart priced a human life at $7 million) argue that the ultimate value of a sense of security is impossible to reduce to a dollar figure. But Mueller’s point is that we have to do these calculations precisely because terrorism is so politically and emotionally fraught. There are lots of things that we, as a society, could do to make ourselves safer but choose not to because of the cost, in time or money or restrictions on our freedom—a 25 mph speed limit on the interstate, for example, would save a lot of lives.
With terrorism, Mueller believes, we are unable to think that way, and in a world of finite resources that means money that could be spent more usefully is instead going to feel-good measures that don’t do anything. “What’s your chance of being killed by a terrorist if you’re American?” he says. “It’s now about one in 4 million per year. Maybe that’s enough, maybe that’s not enough. Some people might say we can save some money and make it one in 3.5 million. What I’m trying to do is just apply standard analytic techniques to the hazards of terrorism.”
