Is Aviation Security Cost-Effective?

Since 9/11, the U.S. has spent $6 billion a year on aviation security to prevent a similar attack. The two most direct efforts to prevent airliner hijackings have been the hardening of cockpit doors and increased presence of air marshals on flights. These measures alone have cost the government and airlines $1 billion a year. Is that money well spent?

Levitt has wondered about the costs and benefits of airline security before. Now Mark Stewart, a civil engineer at the University of Newcastle and John Mueller, a political science professor at Ohio State University, have run some numbers.

Their study, which considered the lives of airborne passengers and potential victims on the ground, found that hardened cockpit doors cost roughly $800,000 per life saved. At the same time, they calculate the air marshal program to cost roughly $180 million per life saved (assuming, that is, the marshals aren’t grounded when their names come up on the terrorist no-fly list, a problem the Washington Times reported on earlier this year).

The Federal Aviation Administration considers any innovation which costs less than $3 million per life saved to be cost-effective. By that metric, hardening cockpit doors seems to be cost effective, while the air marshals program is not.

(HT: Bruce Schneier)


I don't know what is or us not cost effective but I know something is broken. On a recent trip between two of our busiest airports, I realized that I had forgotten to leave my tube of hairgel at home, so I removed it from my carryon backpack and threw it in the covenient trash barrel at the head of the security line. Imagine my surprise -- and horror -- to return home and find that an x-acto knife and box of matches were still in my bag and had made it through security not once but twice-- and one of these airports was the origin of two if the 9/11 flights. Something is not working or else there actually is some racial profiling going on and being a white man with blue eyes I was not considered a threat.


Noah has it right, all this money and effort at airport "security" is closing the barn door after the horses got out. The "next" 9/11 won't use the same tactics, that's not the way terrorism works. We're ready for that now. As always governments/military/officials are preparing for the war they just fought and will be caught unprepared for the next attack on a soft target. Which makes it easy to demonize the enemy for being "devious" and not fighting fair. Creativity and realisitic assesment of our weaknesses will help us prepare, spending money and inconveniencing everyone won't.

Joe P.

I don't think there's any racial profiling involved...I think that an X-ray just can't detect a bottle of gel or a book of matches.


@Jim - Nothing is wrong. What could you have done with the x-acto knife and box of matches? Set off the smoke alarm in the lavatory? Cut a few people before the rest of the passengers tore you to pieces? Even if you had a gun, what are the odds that you would make it to the cockpit? The worst that you could do is decompress the cabin, unless you had some serious explosives. Even then, all you could do is crash one plane. A terrorist could achieve a lot more damage using those inside the airport. Another 9/11 is not possible, not for quite some time.


Has anyone calculated the cost per life lost of screening for hair gel?


I'd say the only effective use of money has been on the cockpit doors. Making people feel safe by having them wastefully throw things away and wait in line for minutes/hours does nothing to stop terrorism. All it does is decrease productivity, increase waste and cause unneeded consumer spending.


"They who can give up essential liberty to obtain a little temporary safety, deserve neither liberty nor safety."

Ben Franklin

Would it not be more effective (and cheaper) to have a marshal or two on every flight than other forms of security?

I know this is a bad year for airlines, but costs are rising and ticket price is not, convinces are. Where is the economics in that?

I have not found either government or the airline industry to care about efficiency.


I am no economics expert, but it seems to me that this study neglects another important effect that has a strong impact on cost-effectiveness: how many more tickets are sold as a result of these measures?

As MRB pointed out (comment 1), 9/11 almost bankrupted the airline industry. The losses did not come from the physical damage done by those horrific attacks, but rather from the resulting loss of consumer confidence in the overall safety of air travel. From this perspective, some security measures surely generate revenue for the airlines. In that sense, government safety programs can be considered a type of "stimulus" for the industry.


I think any expense related to "national security" would do badly in a cost-benefit analysis. We don't know how much less probably an attack becomes because of a measure, or even a collection of measures. Yet, billions are spent regardless. Of course, if you question it you're just "weak on terrorism" - parallels to "weak on crime"?

I'd argue it would have been much more effective to invest even a fraction of the anti-terrorism spending on education and economic development in countries where terrorist organizations recruit. You cut off the easy access to new recruits first, then you worry about the ones who are still around. Otherwise it's just an expensive and deadly game of whac-a-mole.


While it is difficult to know what may have happened without cockpit doors, there isn't a single incident on record where a hardened cockpit door has directly prevented a cockpit intrusion.

The hardened door, however, could cost lives in an accident scenario similar to Helios Flight 522 in 2005. The cockpit crew became incapacitated due to oxygen deprivation. A flight attendant was able to enter the cockpit using security procedures shortly before both engines ran out of fuel using a spare oxygen bottle to maintain consciousness.

If the flight attendant had become incapacitated, no one could have entered the cockpit until the engine generator failed, disabling the electronic lock. In a similar situation, an able passenger will not be able to intervene.



Before the attacks - people were told to cooperate with highjackers. Today - those same highjackers would be torn apart. Many of the expensive, ineffective, time consuming measures like ID checking don't do a thing to keep a plane in the air.


How did they work out how many lives have been saved by these measures? How confident can we be in those $800,000/person estimates?


The TSA drives me nuts. I am sick and tired of them instituting policies that make my flying more "secure" and are a huge inconvenience. For example,

the liquid limit. First of all, I have accidentally taken a bottle of prescription gel through security, in my bag, without declaring it.

Finally, is less than 3 ounces really safe? What happens if I have 3 ounces, my friend has 3 ounces, and that guy over there has 3 ounces...

I would prefer less money spent on safety, less TSA regulations making my travel incredibly tedious and inconvenient, and common knowledge that you are not 100% safe when traveling.


I would love to see some data regarding the change in odds of being subject to terrorist attack.

While an attack would be a horrible thing, I think our institutions are vastly over-estimating their ability to prevent it. I'm not sure that lowering the chances of being killed by terrorists from .0001% to .00005% merits the costs, restrictions, and inconveniences (the Yankees prevent people from bringing sunscreen in to the stadium).

What's the cost/benefit on money spent on aid and diplomacy? We've known this in health care for a long time, prevention is more much cost-effective than treatment. I can't help but think we'd all be safer if those billions were spent addressed the underlying cause terrorist intent rather than trying to patch every security hole in society.

Fed Up


The current security regulations are ridiculous! They need to stop the current plan and reanalyze what security precautions are really necessary. I don't have a big problem with not carrying box cutters. I don't have a big problem with them X-raying my shoes. But everything beyond that is SILLY.


As run by the feds? No.


Which rather backs up what is instinctively true. Hardened cockpit doors, good idea, very effective. All other measures taken post 9/11: waste of money.


I didn't read the study but is there really a large enough sample size of terrorist attacks stopped by either measure to find anything statistically significant?


I think it's worth it. 9/11 almost bankrupted the airline industry. The penalty for "another 9/11" (Besides Bush declaring himself Decider-For-Life) would be the end of all the major airlines. So from their standpoint, I think it is worth it.

Not that I think our new security has helped prevent an attack, but the cost of an attack is so catastrophic that I think the money spent is worth it.


Sorry, but I gotta file this one under "Dept. of Numbers Pulled Out of Someone's Ass". In order to get a calculation for cost to save a life, it would have to be estimated how many lives the program has actually saved or will save. This would be pure conjecture.