Using cost-benefit to evaluate aviation security
Stewart and Mueller's paper, "Assessing the risks, costs and benefits of United States aviation security measures," (published by the University of Newcastle, Australia) does an amazing job of unpicking which post-911 security measures actually work and which ones are showy wastes of money and pocket-liners for slimy government contractors:
Hardening cockpit doors has the highest risk reduction (16.67%) at lowest additional cost of $40 million. On the other hand, the Federal Air Marshal Service costs $900 million pa but reduces risk by only 1.67%. The Federal Air Marshal Service may be more cost-effective if it is able to show extra benefit over the cheaper measure of hardening cockpit doors. However, the Federal Air Marshal Service seems to have significantly less benefit which means that hardening cockpit doors is the more cost-effective measure.Link (via Schneier)
the latest
latest episodes
Wild to think that there was no cost benefit analysis done before spending the money.
Wild? This is DHS we're talking about. It's not even remotely surprising.
looking at it the wrong way. Like the Waronsomedrugs Industry, the Warofterror Industry or the Prison Industry or even the Copsnlawyersnjudges Industry (the "legal" system), this business exists to profit the select few and does that well. It IS a success. Just not yours.
Link is timing out. Anyone got a mirror?
I'm confused how they reach the figures they are quoting. How can they say that hardened doors reduce risk by 16.67% (four significant figures!)?
Naturally this is not measuring the decrease in hijackings since adding hardened doors, since there have not been nearly enough hijackings to make such a statistical claim, so I assume they must be making inferences based on assumptions and guesses, which makes the four sig figs even more odd.
(I assume the answer may be in the article, but the link to the article from the abstract page is down).
I was wondering the same thing, SAMSAM. Where do those significant digits come from? In any case, I'm not sure that better cockpit doors and armed air marshals are necessarily solutions to the same problem, despite the handwaving in the summary. A more secure cockpit door is pretty good at stopping a terrorist from actually flying the plane, and that's great. But that's not the only potential attack. Doors won't do anything to prevent someone from murdering the passengers, crippling the plane during takeoff or landing so that it crashes into a building, or simply blowing up the plane, whereas air marshals might prevent those things. (So might arming pilots, and at much lower cost, but it's hard to get a rational discussion of that particular option.)
"I'm confused how they reach the figures they are quoting. How can they say that hardened doors reduce risk by 16.67% (four significant figures!)?"
Easy: somebody said "Oh, I'd guess about one in six," which was then converted to decimal by somebody else who hadn't been paying attention in the first week of high-school physics class.
The Federal Air Marshal program should have been scrapped long ago. It has essentially no benefits and tons of costs (not just the $900MM/year quoted).
Beginning with United flight 93 on 9/11/01, passengers and crews have repeatedly shown that they will not cooperate with hijackers or other terrorists. It was passengers, not FAMs, who prevented that plane from reaching its target. Similarly, it was passengers who kept Richard Reid from lighting his shoe bomb. It was passengers and crew who stopped the Air Mauritania hijacker last year. And there have been multiple incidents where a drunk or mentally imbalanced passenger has tried to open a door or break into the cockpit, been subdued by the passengers and crew, and only after everything was under control did the FAMs on board ID themselves. FAMs aren't needed to deal with hijackers, and their guns are completely ineffective against a real suicide bomber.
Meanwhile, FAMs have shot one crazy guy in the back as he ran off a plane in Florida yelling that he had a bomb. They scrambled F16s and caused a Mumbai-bound plane to return to Amsterdam because a group of Indian businessmen were playing with their cellphones after takeoff. Every day they commandeer hundreds of first class seats for which the airlines receive no compensation. And they're easy to spot, creating a serious risk that a terrorist team could overcome the FAMs on a plane and use the FAMS' weapons against them--weapons that they bring into the aircraft cabin despite all the TSA's absurd efforts to keep weapons (and toothpaste) out of the "sterile zone."
Down with the FAM program!
Rather than running a full cost-benefit analysis, they could also have just asked ten averagely-intelligent adults and the common-sense consensus would have been the same.
If the DHS had been around when people started building houses, would we have locks on our doors, or would everyone have heavily-armed guards lurking on their lawns, disguised as garden gnomes?
been done
http://img155.imageshack.us/img155/355/zimhouse1ev.jpg
You know what'd be fantastic is if we could implement some sort of high-tech system to shock non-compliant travelers into accepting our most patriotic doublethink with a simple push of the button.
Seriously, Takuan @#3 wrote my first thought about this game (obvious as it probably is to most BBers).
16.67% risk reduction? How did they figure this out? How many times did terrorists successfully enter a hardened cockpit?
And also, if the cockpit door is hardened, uh what's left for terrorists to do? Threaten to camp outside the door until one of the pilots leaves to go home?
Federal Air Marshal Service? --- $900 million.
Frightened and subdued population? --- priceless.
@ ANGUSM -
It wouldn't have been obvious to me that FAMs fail a cost/benefit analysis. I'm at a bit of a loss as to why the paper only looks at FAMs and cockpit door hardening - both of them seem like they have reasonable (potential) payoff for minimal inconvenience to airline passengers. I'd be much more interested in a cost/benefit analysis of rules requiring us to submit our flip-flops for x-ray and leave our toothpaste outside security.
Ooh - there's an even better (much broader) cost-benefit analysis of homeland security measures linked from Schneier on security a few days ago:
http://www.schneier.com/blog/archives/2008/07/homeland_securi_2.html
What is the risk here? That terrorist get into the cockpit? What would happen if a terrorist threatens to kill a passenger each ten minutes that the cockpit door is closed? Will a reinforced door work in that case?
I think better foreign policies are more cost effective than marshals, doors or whatever.