July 22, 2008
Cost/Benefit Analysis of Airline Security - "considering their effectiveness, their cost and expected lives saved as a result of such expenditure", are the current air travel security measures worth it? A few yes, several no. What I found most fascinating, however, was the table giving cost-per-life-saved for various security measures. (The table is on PDF page 7 and reproduced within). Banning asbestos costs almost 20x the values lives saved. Banning unvented space heaters and requiring Seatbelt+air bags each cost about 1/70th the value of lives saved. More boring policy analysis within. small>[via Schneier]
If you remember, I was harshing on rear side-impact air bags a while back -- we can ballpark how irrational they are.
Taking some figures from this and this paper by Viscusi (the guy who apparently figures all this out): In 2007 dollars (and dealing in rough figures), the Viscusi paper puts the US statistical value of a life at about $8.4M. From the table below, rear lap/shoulder belts save about $5.4M/life and are worth it.
This NHTSA proposal gives the cost of a statistical life at 4M/life; un-fudging their numbers gives about
about 2.7M/life - 4.5M/life for the high and low end of requiring front side-impact airbags. There's obviously some discrepancy
Finally, compare seatbelt+airbag to rear lap+shoulder belts -- both are rational but the latter is 40x less efficient in cost/life saved. So that's a ballpark for the front vs. rear benefit.
Judging from all that, I think side impact air bags are a giant waste of money, but not as giant as I'd thought. If you have kids it's probably close enough that if you decided you were making the irrational decision anyway I might admit that I would too.
Regulation Year Agency Cost per Life Saved (millions of 1995 dollars) ----------------------------------------------------------------------------------- Unvented space heater ban 1980 CPSC 0.1 Seatbelt/air bag 1984 NHTSA 0.1 Aircraft cabin fire protection standard 1985 FAA 0.1 Steering column protection standards 1967 NHTSA 0.1 Underground construction standards 1989 OSHA 0.1 Aircraft seat cushion flammability 1984 FAA 0.6 Trihalomethane in drink water 1979 EPA 0.5 Alcohol and drug controls 1985 FRA 0.5 Auto fuel system integrity 1975 NHTSA 0.5 Aircraft floor emergency lighting 1984 FAA 0.7 Concrete and masonry construction 1988 OSHA 0.7 Passive restraints for trucks and buses 1989 NHTSA 0.8 Auto side impact standards 1990 NHTSA 1.0 Children’s sleepwear flammability ban 1973 CPSC 1.0 Auto side-impact standards 1990 NHTSA 1.0 Metal mine electrical equipment standards 1970 MSHA 1.7 Trenching and evacuation standards 1989 OSHA 1.8 Hazard communication standard 1983 OSHA 1.9 Trucks, buses and MPV side-impact 1989 NHTSA 2.6 Grain dust explosion prevention 1987 OSHA 3.3 Rear lap/shoulder belts for autos 1989 NHTSA 3.8 Standards for radionuclides in uranium mines 1984 EPA 4.1 Ethylene dibromide in drinking water 1991 EPA 6.8 Asbestos occupational exposure limit 1972 OSHA 9.9 Benzene occupational exposure limit 1987 OSHA 10.6 Electrical equipment in coal mines 1970 MSHA 11.1 Arsenic emission standards for glass plants 1986 EPA 16.1 Cover/move uranium mill tailings 1983 EPA 53.6 Acrylonitrate occupational exposure limit 1978 OSHA 61.3 Coke ovens occupational exposure limit 1976 OSHA 75.6 Arsenic occupational exposure limit 1978 OSHA 127.3 Asbestos ban 1989 EPA 131.8 1,2-Dechloropropane in drinking water 1991 EPA 777.4 Hazardous waste land disposal ban 1988 EPA 4,988.7 Municipal solid waste landfills 1988 EPA 22,746.8 Formaldehyde occupational exposure limit 1987 OSHA 102,622.8 Atrazine/alachlor in drinking water 1991 EPA 109,608.5 Hazardous waste listing for wood-preserving chemicals 1990 EPA 6,785,822.0 Note: 2007 dollars are 1.38 times higher than 1995 dollars.
And of course, wouldn't it be wonderful to have it pointed out that there are about 10,000 causes of death that are more likely than terrorism?
Monthly traffic fatalities are roughly equal to the 11 September attacks. Yet we're willing to readjust our lifestyles to preventing terrorism, while we balk at a 55 mph speed limit.
This is also good: "The Quixotic Quest for Invulnerability: Assessing the Costs, Benefits, and Probabilities of Protecting the Homeland"
1. The number of potential terrorist targets is essentially infinite.
2. The probability that any individual target will be attacked is essentially zero.
3. If one potential target happens to enjoy a degree of protection, the agile terrorist usually can readily move on to another one.
4. Most targets are "vulnerable" in that it is not very difficult to damage them, but invulnerable in that they can be rebuilt in fairly short order and at tolerable expense.
5. It is essentially impossible to make a very wide variety of potential terrorist targets invulnerable except by completely closing them down.
One thing none of the anti-terrorism planners want to admit is that even if you perfectly harden the Airports then terrorists will go after the Shopping Malls, and mutatis mutandis.
I've been thinking about the $6T cost per life saved for the wood-preserving chemicals. There's no way that is the actual cost of that effort (right?) and that one life is saved. But if zero lives are saved then the cost per life is obviously infinite. These leaves only two possibilities that I can think of:
1 - There is some statistical designation for partial lives saved (and how would THAT work?)
2 - Something in their math bumped up against the integer limit of the processor they were using to do the calculations
Is there some other possibility that I'm missing?
It very well might be partial lives saved--when you think about the asbestos thing, it poisons you over the long term if you are exposed to it. The only sensible way to say that asbestos "killed" you is to say that you lived to 55 when you died of asbestos poisioning, when you otherwise would have died at 70. So, the asbestos removed 15 years from your life, meaning that you lost about 21% of your life due to the asbestos poisioning (15/71 [avg. life expectancy at birth]=21%).
I don't know for sure that they're doing this, but the fact that the things that are low on the list are instant killers (seat belts/airbags), while the things high on the list are more carcinogenic and whatnot, leads me to think that they're making some sort of calculation like this.
That's a really good question HC. The fact that it was 7M x 1M totally passed me by.
It is definitely fractional statistical lives; I just noticed the second (more detailed) paper isn't available to hoi polloi, but the other Viscusi paper covers it. (If you want the one paper just email me).
Also: it's differential lives saved; the number of deaths from asbestos by banning it and removing it vs. the number of deaths if you don't. Sure, people die from poisoning, but tamper-proof seals on cold drinks don't really save lives.
If the chemicals were mostly harmless; and making it onto the list causes all sorts of cumbersome OHSA regulations; then everything with wood preserving chemicals might get a lot more expensive... For instance, the wood to build every house constructed in the US?
The implications of that could be: since we're around wood with those chemicals anyway, IF they're hazardous then sequelae of improper waste disposal is the least of our worries. Just as with the tamper-proof seals, perhaps there's a correlated risk, and a lot of effort targeted to a statistically insignificant contingency.
« Older Bee's Needs | Baby's First Internet Newer »
To post comments to a thread you must login or create a profile.
It would have been nice to see the cost and expected lives saved per year on these numbers. I'm sure that the 7 trillion number is just because noone is expected to die from wood-preserving chemicals being unlisted, but how much was actually spent would be a good thing to know.
In terms of airport security the article does a good job in stressing that the 9/11 attacks have made passengers more likely to fight back, which is a security measure that costs nothing. All of the measures that have been instituted are reactionary, but future terrorists know that thing that have been tried before are met with more resistance the next time, so likelyhood of an attack along that vector is reduced. It would also be interesting to put the annoying measures (liquids/ shoe removal) in terms of productivity-hours lost.
posted by pino at 04:38PM CST on July 22