Asymmetrical Information - Megan McArdle

06.13.13

We Shouldn't Treat Terrorism the Way We Treat Bathroom Falls

Not all threats are equal, and numbers aren't the only important difference.

My friend Conor Friedersdorf had a typically thoughtful essay last week in which he argued that we are giving up too much liberty to fight terrorism.  Terrorism is a tiny threat compared to many more ordinary causes of death:  

Of course we should dedicate significant resources and effort to stopping terrorism. But consider some hard facts. In 2001, the year when America suffered an unprecedented terrorist attack -- by far the biggest in its history -- roughly 3,000 people died from terrorism in the U.S. Let’s put that in context. That same year in the United States: 71,372 died of diabetes. 29,573 were killed by guns. 13,290 were killed in drunk driving accidents

I am broadly in agreement with Conor that we are giving up far too much liberty to fight terrorism, with the NSA's incredibly broad data-collection programs being only the latest example.  To add insult to injury, much of what we're doing isn't very effective--Americans are still taking off their shoes in every airport line because of a failed attempt to use shoe bombs ten years ago.  X-raying and partially undressing its citizens millions and millions of times a year in order to foil a plot that was foiled not by the TSA, but by passengers who wondered why this guy was trying to set his shoes on fire, is beneath the dignity of a great nation.

Jeffrey Goldberg, however, pushes back on the common libertarian tactic of comparing terrorism deaths to ordinary risks, and arguing from there that we basically shouldn't try to do much about terrorism:

We know for certain, however, that al-Qaeda, its offshoots, and other organizations and individuals in the Islamist orbit seek unconventional weapons that would allow them to kill a far-larger number of Americans than died on Sept. 11.


As early as 1998, Osama bin Laden asserted that Islam required him to use weapons of mass destruction in the conduct of his jihad, and he made the acquisition of these weapons a high priority. The al-Qaeda leader Sulaiman Abu Ghaith famously argued that Muslims had the right to “kill four million Americans, including one million children, displace double that figure, and injure and cripple hundreds and thousands.” 


Is there anyone who actually believes that al-Qaeda or its offshoots would hesitate to use chemical or biological weapons against Western targets if they could? The only reason radical Islamists haven’t used such weapons is that they haven’t been able to acquire them -- mainly, I think, because of effective American countermeasures.

Goldberg is right: you can't look at the low level of terrorism now, and conclude that it's not worth spending much money or even liberty in fighting it.  Maybe if we weren't doing all this stuff, there would be loads and loads of terrorist attacks.  I'd argue that given the small number of attacks foiled on US soil, our choosiness about issuing visas is doing much of the work, and the terrorist fixation on doing stupid things to airplanes is doing much of the rest.  But you need to make that argument; you can't just say, well, we have so few terrorist attacks a year that terrorism must not be a big threat.  

The problem is, it's really not appropriate to compare accidental deaths to homicides, because accidental deaths are not deterrable.  

I'm not saying they're not controllable: better roads, better car design, better warnings about bathtubs can all lower the rate of deaths from accidents.  But a public awareness campaign about accidental drownings is not going to change the behavior of the bathtub.  

Terrorists, on the other hand, are deterrable.  If there's a very high chance they'll get caught before carrying out their mission, they will be less interested in carrying out their mission.  Especially if they know that getting caught means that the full imagination of the society they've targeted will be put to the task of making their lives miserable.  

That's why human beings are evolved to overreact to certain kind of threats.  We worry more about getting mugged than about having our bank rip us off with hidden fees.  We find rape more traumatic than gonorrhea, even though the physical effects of the STD may be much worse.  And we fear terrorism more than non-steroidal anti-inflammatory drugs such as aspirin, even though the latter kill many more people every year than the former.

It makes evolutionary sense that we're hardwired to go postal about intentional harm from other people.  This is what provides the deterrent.  Terrorists and other criminals know that we spend far more time looking out for these sorts of threats, and that our reaction to anyone who is caught doing this sort of thing will go far beyond a rational cost-benefit calculation.  The human instinct for hypervigilance and hyperpunishment undoubtedly means that we have much less crime than we otherwise would.  

That's why I get uncomfortable when libertarians start comparing terrorism to diabetes.  I used to make those sorts of arguments, but I don't any more, because I've come to realize that these arguments miss the key element of deterrence.  It's using a snapshot to analyze something that can only really be captured in a film.