Risk Analyst Refuses to Take A Risk

Brian Doherty | August 21, 2007

The risk, that is, of violating the taboo against suggesting that maybe we aren't all about to die, die, die!! I tell you at the hands of the enemy out there, a-sleepin' and a-plottin'.

Bryan Caplan notes on his blog that Kip Viscusi, the great scholar of risk assessment, refuses to apply his own intellectual standards to the topic of the risk of terrorism and admit that, near as we can tell, it ain't much.

Caplan is the author of the cover story in the forthcoming October print issue of reason (subscribe now!), excerpted from his much-praised new book The Myth of the Rational Voter.

I asked "Where's the Terror?" last September.

Comments

Tim|8.21.07 @ 2:38PM|#

We need to take all precautions to defend against a possible terrorist strike or our entire civilization will be imperiled. Even one death is too many. I have much more to say, but it's hard to type while I'm stuck in traffic and I've just got to take this call.

|8.21.07 @ 2:52PM|#

The Book: One of the many things Ford found most difficult to understand about human beings was their inexplicable habit of stating and restating the very very obvious. As in: "It's a nice day." or "You're very tall" or "So this is it. We're going to die."

|8.21.07 @ 6:02PM|#

It's beyond doubt that Homeland Security (not to mention the Iraq war) is wasteful and extravagant. Still, some of Mr. Viscusi's reticence is justified by the an essential difference between the terrorism threat and, say, the threat of a drought or a cancer. That is, droughts and cancers don't look to see if you are doing anything to detect them. Droughts and cancers don't react to preventive attacks on their finances or their bases. Droughts and cancers don't get discouraged and give up. Terrorists do.

stephen the goldberger|8.21.07 @ 6:43PM|#

piper tom just said everything i was going to, other than to emphasize that all citizens expect their gov'ts to protect them from foreign attacks, not from drought or cancer (at least not most of them...i hope).

brian|8.22.07 @ 10:01AM|#

stephen the goldberger
all citizens expect their gov'ts to protect them from foreign attacks, not from drought or cancer (at least not most of them...i hope).

Thought experiment: why not? what's the difference that puts one in the realm of goverment and the other outside it?

note i'm not making an argument here, just asking a question.

Ken Hagler|8.22.07 @ 1:41PM|#

I'm sad to say that here in the country that was once America, most people do indeed expect the government to protect them from drought and cancer.

|8.22.07 @ 11:41PM|#

PiperTom lays out the argument well. I'd add that "quantifying" the risk of terrorism (and moreover, putting money on the line) is a very different thing than offering blanket assessments like such as "it's likely/not likely"; "it's a big problem/small problem"; "we spend too much/we don't spend enough."

I share Bryan's academic interest in what a prediction market might look like, and I agree with him that the moral hazard issue is basically overblown. But on the question of whether the risk itself is quantifiable, well, the empirical evidence is against him. After 9/11, property insurers and their reinsurers acted en masse to explicitly exclude terrorism from their contracts. Those offering commercial coverages and workers' compensation returned to the market only because they were compelled to by the TRIA law, which also set up a $100 billion government backstop for losses. The reinsurers still have not returned, and in lines of business not covered by TRIA, such as homeowners, there remains no coverage for terrorism. And the reasons mirror the ones piperTom suggests -- it is the ultimate low-frequency, high-severity event, which means actuarial data is useless. You can model the severity of any number of potential scenarios, but there is no upper bounds on the number of possible scenarios. And even if you could model all the possible scenarios, it would tell you nothing about frequency. Without those basic inputs, calculating rates -- which is to say, "quantifying the risk" -- is just shooting blind.

TallDave|8.24.07 @ 1:45PM|#

Why are the people whining about the "politics of fear" and calling anyone who takes terrorism seriously a bedwetter also the same people who claim we're all going to die from global warming?