Bryan Caplan  

Free Market Airport Security

PRINT
Would the Private Sector Make ... Prosecutorial Persecution...
I think Garett's basically wrong about airport security on the free market.  Yes, both markets and politics respond to risk misperceptions.  But the political response is much more likely to ignore cost and convenience, to impose whatever sounds good.  The market response will still pander to misperceptions, but it will weigh these misperceptions against cost and convenience.  Both markets and government embrace slogans like "You can't put a price on safety," but only governments ever take such slogans seriously.

If this seems too abstract, here's one massive cost-cutting, convenience-raising change I'd predict in a free market: Profiling.  Private security firms would still claim to treat everyone equally.  But they'd wave the elderly, women, families with children, and well-dressed men right on through, especially if they look like "regular Americans."  Private security firms would redirect their spare attention to (a) young men who look like they might be violent criminals, and (b) anyone who looks like they might be of Middle Eastern descent.  Free market airport security would arguably be less fair than the status quo, but the total cost would be markedly lower and average convenience would be markedly higher.



COMMENTS (15 to date)

There's a natural experiment going on right now which casts Garett's view into doubt.

Several airports in the US use private contractors, instead of TSA agents. SFO is one of them. These airports consistently rank higher on convenience and courtesy measures and the private contractors do a better job of spotting potential threats.

In response, the TSA has closed the program that allows airports to opt for private security.

matt writes:

Hmm...Wouldn't they get sued for discrimination. Companies do all sorts of weird and ineffective things to avoid being sued. Do you think they would be more effective at avoiding bad publicity from this than governments. What happens when someone wiki-leaks their training material?

John Thacker writes:

The biggest difference is that the public system give everyone a vote, but a private system would give only people who fly a vote, with more votes for people the more that they fly. I believe that surveys I've seen suggest that the more people fly, the less they believe in the extra post-9/11 intrusive security.

Ken B writes:

So much for one commenters snark that Libertarians won't talk about this because of the profiling issue. He's often wrong but Bryan is always fearlessly honest.

I was visiting England. The ONLY rude attendants we met were at the courts. They were very rude indeed. Nowhere where I paid to get in was anyone rude, and I still had to open my bag.

Jim B writes:

I disagree with the conclusion that a market would lean towards profiling, mainly because profiling is not a great solution from a security and even a cost perspective. See Bruce Schneier's arguments here. Like so many things, it is easier and more effective to apply a simpler process to everyone. My guess is a market would implement pre-9/11 style airport security, realizing that you won't get better results for the investment (and that, as Schneier states often, the most effective mesaures against another 9/11 style plot were reinforced cockpit doors and passengers that will fight back). I could also see an argument for a PreCheck type program being encouraged by a market system, but that is not the same as airport checkpoint profiling.

Floccina writes:

Many people die in car accidents but the car companies are not putting governors at 50 mph in new cars nor are new cars built like tanks. For about $3000 more they could be as safe as race cars but they are not. Based on that I think security would not be as inconvenient if run by the airlines but they might beg the Gov to be the heavy for them.

Greg writes:

What odds would you put on the next attempted bombing of an American airline to be conducted by someone who looks Middle Eastern?

Fonzy Shazam writes:

I believe Floccina is on to something in the last sentence of the comment: "[airlines] might beg the Gov to be the heavy for them". It is likely the airlines would not want to leave a system where the government provides the overkill on security, picks up the tab, and is a deflection point for overall customer dissatisfaction. If they had to, they probably would want lots of mandated security measures to both limit competition and provide cover for customer-unfriendly policies.

I do think a private system would be preferable (I side with Bryan). My guess is that the evolution in security would result in a two-tiered system with airports providing one layer and airlines another. In this way airports would have airlines as the direct customer. Airport security would be more stringent than airline security both because it is the first line of physical defense and because the airlines would find this convenient. The totality would be better than the current TSA model, and that system would be more likely to adapt in beneficial ways (both more effective security and less costly security).

Glen writes:

At the very least, private airline security would certainly tailor their invasiveness to the wishes of their best customers. TSA instead takes a “zero tolerance” approach that attempts to (but can never successfully) achieve perfect security. Only the rationally ignorant average voter and/or very infrequent flyer would prefer such service.

Jehu writes:

Perhaps a private airline doing its own security might issue every passenger a short sword and tell them that, in the event of a terrorist attack, they are instructed to either prevail or to die with honor. Maybe the airline instructional video could be narrated by an appropriately martial band of diverse characters.

yarbel writes:

Two points:

a) Profiling is not very effective in preventing future large-scale terrorists attacks, because any non-random auditing mechanism is subject to strategic behavior on part of the terrorists.

b) What's really seem to me to be the heart of the matter is the externality that's taking place. If you get singled out and taken out of the line for a security search, you suffer some social harm (people will look at you badly, you feel embarrassed, etc.). I think that the contention is that private security companies will not account for this externality, whereas the government does. It may be the case that either the government doesn't or that it places too much weight on this externality, but that's an argument that needs to be made.

(for the sake of disclosure, I am a middle-eastern man...)

Chris H writes:

yarbel writes:

a) Profiling is not very effective in preventing future large-scale terrorists attacks, because any non-random auditing mechanism is subject to strategic behavior on part of the terrorists.

There are two important caveats here. This only works if terrorists a) know the profiling method and b) can actually respond to it.

For the first caveat, this might actually tend to reduce basic age/race/gender profiling as these types of profiling are hard to hide. But more complex statistical analysis of flight and ticket buying patterns, types of movements within the airport, and potentially numerous other variables security experts might come up with. Figuring out these profiling methods might be far more difficult and thus defeat attempts at strategic avoidance by terrorists.

The second caveat though may be even bigger. Only a miniscule fraction of young Muslim Arab and black African men ever sign up with terrorist organizations, much less for highly dangerous or suicide type attacks. Getting other ethnic groups and genders is likely to be far more difficult. Attempts to coerce people into terrorism would be highly problematic given the nature of suicide attacks ("I do a suicide attack or you kill me? How's that a choice!" even threatening family members is not guaranteed to work), thus meaning that in nearly every case the terrorists in question will have to be volunteers. If young Arab and black African Muslim men are extremely unlikely to sign up for such missions, how much more unlikely are elderly white females? Finding volunteers outside profiled demographics may be so difficult as to make terrorist attacks unfeasible.

But there is your other point to consider:

b) What's really seem to me to be the heart of the matter is the externality that's taking place. If you get singled out and taken out of the line for a security search, you suffer some social harm (people will look at you badly, you feel embarrassed, etc.). I think that the contention is that private security companies will not account for this externality, whereas the government does.

I think the problem here is what your talking about is not an externality. Consider that the prospect of suspicion and embarrassment will reduce the number of profiled customers to the airport engaged in profiling. If profiling is a service valued by consumers then airlines will be able to raise prices correspondingly meaning airline consumers will pay higher prices for receiving this benefit. If airline consumers do not believe they benefit from this service, then in order to profile airlines will have to accept lower profits due to lost business or due to price decreases for profiled customers to keep their business. The result will tend to mitigate or even eliminate profiling.

What we have here is not an externality but a cost fully internalized by either the profiled person deciding to undertake flights, the broader customer base of the airline willing to pay higher ticket prices, or the airlines willing to forgo higher profits. In cases where externalities do not exist, there is no logic for a government intervention and the government should not be considered superior at how to manage the situation.

norse writes:

Well, given that a system in which there is no control would be just as effective... and think about how much we could reduce costs!

http://marginalrevolution.com/marginalrevolution/2012/11/the-high-price-of-false-security.html

rapscallion writes:

Ken B,

Bryan is a welcome, honest and forthright exception to a tendency.

Carl Jakobsson writes:

What about passengers having different preferences for various forms of security-measures? If the market is divided, wouldn't some airlines have stricter controls and others less so?

Comments for this entry have been closed
Return to top