Privatizing Airplane Security

In a column earlier this week, John Tierney of the New York Times suggests that current approaches to safeguarding airplanes are ineffective because the Transportation Safety Authority is bogged down with bureaucratic inertia and misfocussed Congressional mandates.  The particular issue Tierney highlights is whether screening should consist of searching bags or of identifying people who behave suspiciously.   His overall concern, however, goes beyond this one issue:

It’s not that the T.S.A.’s leaders don’t see the problem. Kip Hawley, who took over the agency last year, is a smart manager who has been trying to change the agency’s focus. He removed small scissors from the taboo list, and he has complained about all the time spent by screeners seizing cigarette lighters to comply with an order from Congress.

But he’s making little headway because he has inherited an unworkable mess created by Congress after Sept. 11. It ignored the security model in Israel and much of Europe, where screening programs are run by airports under the guidance of a national agency. Instead, Congress ordered the T.S.A. to both supervise and run the screening programs itself.

The result has been a waste of billions of dollars on an unwieldy federal agency that’s become known as Thousands Standing Around. The T.S.A. should be trying to anticipate new terrorist tactics, like the bomb plot uncovered in England, but it had to raid its research budget to pay for the screening program, as Eric Lipton and Matt Wald reported in The Times.

I do not know whether Tierney and the sources he quotes are correct about screening bags versus screening people, but I do think there is only one way to find out:  get the government out of the airplane security business and leave security matters to the airlines.

Now, this suggestion will strike many as lunacy.  A standard assumption is that without government provided security measures, there will be no security provisions.   Airlines, so this thinking goes, are only interested in profits.  Security is expensive, so airlines will undersupply it to the determinent of their passengers and anti-terrorism efforts more generally.

In fact, airlines face strong incentives to make their planes safe.  Consumers will refuse to fly on airlines that do not provide adequate safety.   Pilots and flight attendants will do the same.  And insurance companies will not insure airlines that fail to take adequate security precautions.

The real question, therefore, is whether a privatized system would be more effective than the current system?  What exactly might a privatized system do differently?

The short answer is, we do not know.  A crucial negative of government imposed "solutions" is shutting down the innovation and experimentation that would occur if the private sector were free to act.

But it is easy to imagine approaches that airlines might adopt if they could provide security in whatever manner they saw fit.

Some airlines might develop preferred flyer lists that allow individual passengers to submit to extensive background checks and then become eligible to board planes without waiting in long lines.  The TSA has considered this approach, but so far it has not been implemented.

Some airlines might decide that the best approach is complete separation of passengers and baggage.   That is, you fly on one plane while your bags fly on another, thereby elminating the risk of bombs in either checked or carry-on bags, at least for planes with many people on board.

Some airlines might utilize extensive air marshal programs.

Some airlines might utilize extensive profiling.

But what airlines would not do is spend countless hours screening people who pose no risk.  They would not expend enormous resources screening checked luggage even while carrying cargo (packages sent by third parties who are not traveling on the planes) that is totally unscreened.  And airlines would not pay inflated union wages to employees with minimal skills or training.

To be clear, no system can be entirely safe.  A determined terrorist will occassionally find a way to beat any system.

But privatization will not only provide information about what works and what does not; it will do so in a far more efficient manner than the TSA and the government imposed system it operates.

Comments

[Mike Huben]

There are four really obvious problems with this.

First, airlines have ALWAYS been able to augment security. There is no government regulation stopping them. Indeed, the Captain of a plane can refuse to transport any passenger he deems suspiscious. So we have little reason to believe that a privatized system would be any better.

Second, the problem is not simply the direct responsability for the airplane and crew, but the externalities (such as the twin towers, international tensions, and investigative costs.)

Third, proper security would require private access to national security databases: that's not going to happen.

Fourth, non-uniform screening would simply encourage terrorists to work through the airline with the poorest screening, and cost cutting due to competition would likely present even more opportunities than exist now.

Maybe Miron should take a trip across the river to Harvard Business School and consult with David A. Moss, author of When All Else Fails: Government as the Ultimate Risk Manager. Kneejerk privatization suggestions might sound good for about 15 seconds, but quickly show their faults.

Just re-raising such suggestions a little later shows how silly they are. For example, in Libertarianism in One Lesson; The Second Lesson, I bring up a classic: "Harry Browne had it right in his 2000 campain. Trust to the efficacy of the market! As soon as you threaten to put a bounty on the head of a terrorist, all your terrorism problems will be solved."

[Andrew Leigh]

Jeff, I think the negative externalities would be the hardest problem for this to counter. Suppose that I set up a "no security" airline. It's so cheap to run that it offers tickets at half the price of Southwest. But when my planes get bombed, there's a negative spillover on other airlines, and via increased stress on non-fliers. Have you thought about ways the exernality could be internalised? Fine them if they get bombed, perhaps?

[James]

Mike & Andrew, if airlines had to bear the cost of any damages from terrorism made possible by their negligence, I suspect that they might see that as a reason to take security seriously.

Mike, I'm not sure that airlines are really free to do whatever they wish without a regulation to stop them. If you consider antidiscrimination laws to be regulations, they certainly can't choose to augment existing security by e.g. subjecting all members certain demographic groups to searches above and beyond what TSA conducts.

I think we have at least some reason to believe that a privatized system would do a better job. This article seems to indicate that much. At any rate, the situation we have now makes a real empirical comparison impossible. There are probably some statists so dishonest as to argue that the relative lack of evidence should be interpreted as a reason not to privatize, i.e., don't try anything until we know that it's been tried and has worked. I'm sure you're not among them.

Another problem with the present system that Miron left out is the perverse direction of redistribution. The current approach uses tax dollars from everyone, including those who never fly, to subsidize the security costs affecting the small but affluent minority that use air travel on a frequent basis. This not necessarily a reason to favor privatization, since the government could pay for security by charging a security services fee to airlines. But privatization would also be an effective way of ending this form of redistribution from the middle to the top.

[David]

Airlines did provide passenger screening and security themselves, prior to 9/11. That provision was subject to certain requirements from the government, but airlines were free to subcontract the screening function to other firms, which they largely did.

James is not quite correct, since security is at least partially funded through a specific September 11 fee of $2.50 per one way journey. This fee is paid even if a ticket is acquired with frequent flier miles. Other funding comes from a federal assessment on the airlines themselves (the Aviation Security Infrastructure Fee, or ASIF), which is partially based on the estimated amounts that airlines themselves spent on security functions prior to 9/11.

[James]

David,

You're right about the funding, but unless the fees you mention cover the entire cost of securing flights, there is still some upward redistribution going on. As I mentioned before, however, I would see a fee structure where security funding comes entirely from air travel to be an adequate way of dealing with this particular issue.

I'm aware that the airlines were free to use private security in the past. But I'm somewhat skeptical of empirical approaches to the issue for a few reasons. For one, terrorist attempts are very rare events. It would take too much time to build a statistically valid sample. For two, the government would have to actually allow privatization in order to yield meaningful data on the consequences of privatization. Otherwise, we'd be left attempting to judge a free market from the outcomes of a regulated market.

[David]

James -- the fact that direct funding of air security is partial is why I said you were not quite correct. Whether the consequences of successful terrorist events involving air carriers are limited to the victimized airline, or extend to the broader economy, is a relevant question when deciding about these funding sources -- perhaps some general fund participation is a more accurate reflection of the actual range of consequences.

In addition, the federal government in effect immunized airlines and other parties against tort claims by 9/11 victims as part of the whole sorting out and settlement process following those attacks. Even if those airlines (American and United) had been financially healthy, they could not have covered the likely judgements against them.

Finally, to the extent that effectively thwarting terrorist actions involves much much more than stopping nasty people at the screening point, the problem of coordinating private sector and public sector actions and incentives in this area is much more complex than what might fit into the rather tidy model you seem to be working from.

[James]

David,

Since it doesn't seem to me that your first two paragraphs actually deny anything I've said, I'll only respond to the third.

Yes, the world is complex. This doesn't necessarily imply that simple models are a problem. In many applications, this may indicate that simple models are the best approach. Each additional assumption is another potential point of failure that is liable to break down as a result of the complexity of the process being modeled. So I try to limit myself to three fairly safe assumptions: that people respond to incentives, that things tend to work better when the legal system internalizes externalities, and that the incentives faced by the persons in a nationalized industry are difficult to align with the intended goal of that industry, especially if the people in that nationalized industry can force others to buy what they are selling.

I'll be the first to admit that I may be leaving something out, but "you might be oversimplfying" is not a coherent argument for central planning. (I'm not accusing you of making such an argument. It's just a common one.) Do you have a model that accounts for some complexity that my simple model ignores?

[David]

I've already noted the things your "simple model" is leaving out: the history of airline security provision, the facts of existing aviation security funding and the externalities associated with nasty aviation security events. Other than that, you've got it covered. haha

[James]

David,

I also left out the hair color of the pilot. But it's on you to show that adding in assumptions about these extra variables makes the model more informative, rather than less robust. Or you could flippantly assert it and laugh, I suppose.

[Mike Huben]

James, you are accusing others of your own sins. In order for your model to be simple, YOU are making assumptions that there are no significant externalities and incorrect assumptions of aviation security funding. David is merely pointing out YOUR assumptions.

Really, you couldn't look sillier. And I note that you haven't adequately addressed any of my 4 points. But of course, Miron hasn't either.

[James]

Hi Mike.

Recall where I wrote "I try to limit myself to three fairly safe assumptions: that people respond to incentives, that things tend to work better when the legal system internalizes externalities, and that the incentives faced by the persons in a nationalized industry are difficult to align with the intended goal of that industry, especially if the people in that nationalized industry can force others to buy what they are selling."

In other words, if there are externalities, the best approach is to internalize them. Since the world is complex and the actual externalities may change, I didn't want to clutter things by making more specific assumptions about what the externalities are. Still think I'm neglecting externalities?

It maybe that David was pointing out what he saw as poor assumtions that I had made. I wish he'd said so. When he used the word "tidy" I wasn't sure what he actually meant and guessed that he was trying to say that I had too few.

Well, if you'd like for me to address your four points that's a fair request.

1: Observe what I've alredy written here. Airlines are not entirely free in how they may augment security and captains are not free to refuse passengers at will so long as the threat of lawsuits exists.

2: Yes, externalities exist. This alone is not sufficient to show that privatization or nationalization is better. However, we know that externalities tend to be a larger problem in a commons scenario than in the context of strong property rights. Since a nationalized industry creates a security commons, I'd be surprised if externalities were actually reduced by having a nationalized industry.

3: You assert that security would require access to national databases. I couldn't begin to address this until you tell me why you believe it, and exactly which databases you mean.

4: I could address this more effectively if you show me the asumptions and actual steps, first order conditions, etc., you used to derive this result and why it would be bad.

Otherwise, it just seems like you are asserting it and expecting me to take it as given. I could just as soon assert that airlines would start competing on the basis of security rather than cost, but I won't.

Here's my guess: airlines would produce the bundle of security and cheapness that best satisfies the demands of their consumers, i.e. setting the marginal benefit of security equal to the marginal cost. This may lead to decreased security, but I don't see that as automatically bad. The present arrangement may be well to either side of the point where the marginal benefit and marginal cost of security are equal.

[Mike Huben]

"I usually make a fairly safe assumption that the sky is blue. Because the sky is a commons, I expect that it is not blue as often as it would be if we had strong property rights. You assert that the sky is gray today and black at night: I can't begin to address this until you tell me why you believe it and exactly which databases say this. I could address this more effectively if you show me the asumptions and actual steps, first order conditions, etc., you used to derive this result and why it would be bad."

Somebody give that man a PhD in pompous sophistry!

[Eric H]

Huben is right on his second point, but wrong on his first, and 3 and 4 are both moot when making comparisons to the status quo. It's fun to invoke Second Best theory and to stop all analysis at that point, but not very convincing.

[Mike Huben]

Considering that Miron is astute enough to know Second Best theory, we should be convinced that he's selling us a bill of goods when he improperly uses simplistic microeconomic arguments.

[James]

Huben,

Asking a person to defend their assertions is not usually considered pompous. I suppose it's easier to be flippant than to admit that you were making unfounded predictions.

[Pharadei Marts]

All preparat FDA aproved:
Acyclovir ,
Ambien ,
Amoxicillin ,
Ativan ,
Celebrex ,
Cialis ,
Cialis Soft Tabs ,
Glucophage ,
Lasix ,
Levitra ,
Lipitor ,
Meridia ,
Paxil ,
Propecia ,
Prozac ,
Soma ,
Valium ,
Viagra ,
Viagra Soft Tabs ,
Xanax ,
Zoloft .

[juliavanderbat]

Hello


Phentermine is indicated only as short-term monotherapy for the management of exogenous obesity.
More about of phentermine - The best choice Phentermine (http://the-bestchoice.blogspot.com/ )

Bye

[Shawny]

It would be stupid for the US, Govt. to allow a private security company to screen bags and people. The Airports will compromise passenger security for the lowest security contract. The standard of security would drop due to under paid security guards. In addition, drugs and other illegal items may get through with a private security guard staff. Private security should never screen baggage, just patrol the airport grounds parking lots.

[dellaseymo]

signed possibly taken simulations dimming references research frequency