Schneier on Security
A blog covering security and security technology.
« My New PGP/GPG and OTR Keys |
| Government Secrecy and the Generation Gap »
September 9, 2013
Excess Automobile Deaths as a Result of 9/11
People commented about a point I made in a recent essay:
In the months after 9/11, so many people chose to drive instead of fly that the resulting deaths dwarfed the deaths from the terrorist attack itself, because cars are much more dangerous than airplanes.
Yes, that's wrong. Where I said "months," I should have said "years."
I got the sound bite from John Mueller and Mark G. Stewart's book, Terror, Security, and Money. This is footnote 19 from Chapter 1:
The inconvenience of extra passenger screening and added costs at airports after 9/11 cause many short-haul passengers to drive to their destination instead, and, since airline travel is far safer than car travel, this has led to an increase of 500 U.S. traffic fatalities per year. Using DHS-mandated value of statistical life at $6.5 million, this equates to a loss of $3.2 billion per year, or $32 billion over the period 2002 to 2011 (Blalock et al. 2007).
The authors make the same point in this earlier (and shorter) essay:
Increased delays and added costs at U.S. airports due to new security procedures provide incentive for many short-haul passengers to drive to their destination rather than flying, and, since driving is far riskier than air travel, the extra automobile traffic generated has been estimated in one study to result in 500 or more extra road fatalities per year.
The references are:
- Garrick Blalock, Vrinda Kadiyali, and Daniel H. Simon. 2007. "The Impact of Post-9/11 Airport Security Measures on the Demand for Air Travel." Journal of Law and Economics 50(4) November: 731–755.
- Garrick Blalock, Vrinda Kadiyali, and Daniel H. Simon. 2009. "Driving Fatalities after 9/11: A Hidden Cost of Terrorism." Applied Economics 41(14): 1717–1729.
Business Week makes the same point here.
There's also this reference:
- Michael Sivak and Michael J. Flannagan. 2004. "Consequences for road traffic fatalities of the reduction in flying following September 11, 2001." Transportation Research Part F: Traffic Psychology and Behavior 7 (4).
Abstract: Gigerenzer (Gigerenzer, G. (2004). Dread risk, September 11, and fatal traffic accidents. Psychological Science, 15 , 286287) argued that the increased fear of flying in the U.S. after September 11 resulted in a partial shift from flying to driving on rural interstate highways, with a consequent increase of 353 road traffic fatalities for October through December 2001. We reevaluated the consequences of September 11 by utilizing the trends in road traffic fatalities from 2000 to 2001 for January through August. We also examined which road types and traffic participants contributed most to the increased road fatalities. We conclude that (1) the partial modal shift after September 11 resulted in 1018 additional road fatalities for the three months in question, which is substantially more than estimated by Gigerenzer, (2) the major part of the increased toll occurred on local roads, arguing against a simple modal shift from flying to driving to the same destinations, (3) driver fatalities did not increase more than in proportion to passenger fatalities, and (4) pedestrians and bicyclists bore a disproportionate share of the increased fatalities.
This is another analysis.
Posted on September 9, 2013 at 6:20 AM
• 42 Comments
To receive these entries once a month by e-mail, sign up for the Crypto-Gram Newsletter.
Weirdly enough, my dad says a lot of pedestrian fatalities are bus-related, ie: a bus turning a corner and hitting you. You would think people would turn their head...so now I have a crosswalk-system w/ an annoying robotic voice telling me to "Wait...Wait...Wait..." when there's no cars/buses in sight and I get visions of big brother.
Another good illustration:
In the past 20 years total, deer have killed more on American soil (~150-200/yr) than terrorists.
So where is the trillion dollar War on Bambi?
Those crosswalks are to assist blind and visually impaired people crossing the street.
Wasn't there also a claim that since the DHS took over things like food safety and redirected agents away from inspecting food plants to confiscating 2.1oz bottles of water - that there had been a large rise in food poisoning outbreaks ?
So, in essence, one of every sixty four traffic fatalities is a traveller who is driving because of post-9/11 airport security theatre?
Overall, airline passenger miles are up during the period, and highway miles are down (the later perhaps due to economic issues). Road fatalities are also down.
I'm not convinced.
There are several confounding variables here that have not being considered. Throwing out numbers like that doesn't help anybody.
This example is one that makes me think that something is missing in the traditional assessment of risk assessment. I know the narrative of human cognitive biases and responding disproportionately to rare events, but I think that there is another effect where it comes to risk combined with a sense of agency. People accept automotive deaths because there is the perception that we all have the power to limit this risk through the way we drive. Airline crashes, terrorism and even mundane things like food and drug safety are places where we feel that we have no agency to prevent the risk so those persons that have the power, be they corporate or political leaders, are held accountable. This is probably the reason that drunk driving became less acceptable as said drivers began to remove personal agency from driving safety as drunk drivers tended to kill others no matter what they did.
We can call treating risk that we accept and risk that we don't accept differently irrational and that is where the whole issue of risk discounting can come into play, but I feel from a psychological level the issue involves agency and not people just ignoring certain risks outright.
Just to make things worse there is a counter argument which I'm not even going to try to find figures for due to too many variables and insufficient recording of information.
It's all to do with doses of radiation and the effect that such ionising of your DNA has on your life expectancy.
It is known that ionizing radiation increases with altitude and there is some evidence that airline flight staff have an increased level of radiation related diseases and thus in some cases a shortend life expectancy.
There is also an argument that body scanners might increase this risk (currently there is no data to support for/against/no effect for obvious reasons).
So some have argued that those who drive will get a lower life time dose of ionizing radiation and thus have an increased life expectancy because of it.
Personaly I'm not going to argue the case either way, I'm simply pointing out there are other arguments to consider over all, and along those lines are other arguments such as,
Aircraft have poor air quality (due to costs involved) and this has all sorts of health implications, due not just to worsening existing cardio-vascula issues but to disease transfere such as TB and SARS etc.
Further passengers are these days are effectivly "locked in" to their asigned seat and often activly discoraged from leaving it to get excercise etc by aircrew and other passengers. For many years it has been stated that this increases the risk of DVT, PE and CE, which can prove to be either permanently debilitting (Stroke) or fatal within a very short period after the flight.
Then there are risks...
You get the picture, there's very little definition or contrast and one heck of a lot of noise so the picture is almost impossible to see.
How about all the deaths that could have been avoided if the US government would have invested their tax-payers money into health care and education rather than military and surveillance...
I think the larger point is that traffic fatalities amount to one 9-11 each month, and instead of investing in mass transit and intercity rail, we subsidize more driving.
The Federal Interstate system is heavily subsidized. For the Interstate system to be financially solvent, the Federal gasoline tax would need to be raised by 40¢ per gallon NOW. That's just to make it solvent, not to mention expansion or improvement.
The Federal gasoline tax was last raised by a nickel in 1993, and is NOT indexed to inflation.
This problem is compounded by Federal CAFE standards and tax credits for electric vehicles. People who drive hybrids and plug-in electrics pay less for the roads they use, at a time when people really need to be paying more.
If you buy a Chevrolet Volt, you are eligible for $7,500 worth of Federal tax credits. The Volt is a luxury car. Taxpayers are being asked to subsidize luxury for people who can afford it on their own just fine. Luxury is supposed to come at a premium.
So, instead of treating intercity rail as a national security priority, we have State governors like Scott Walker turn away close to a billion dollars worth of federal dollars to build high speed rail, so that more people drive dangerous, subsidized automobiles.
None of this War on Terror is about saving lives. There are cheaper, less invasive ways of saving American lives (like encouraging people to use transit and incentivising urban living instead of subsidizing suburbs).
The War on Terror is a new Cold War. It's about subsidizing industry.
I'd like to study how accident statistics decompose.
How dangerous is intra-metro area travel compared to inter-city travel? I lived in NYC for many years -- I'd argue that the trip from a home in Brooklyn to JFK, Newark or La Guardia to fly to Boston is more dangerous than the drive on I-95/I-84/I-90 between NYC and Boston or NYC and Washington, DC.
The body count's pretty irrelevant. At least in the eyes of people allocating funds for safety, security, etc., the threat from terrorism is to the legitimacy of the government -- are they doing enough? Were their prior decisions right? Is the government strong/smart/modern enough?
"The Federal Interstate system is heavily subsidized"
...except if you consider what would happen to US productivity without an effective interstate road system.
One can easily argue that every dollar spent on interstate highways is earned-back multiple times in tax revenue.
Well I'm going to call errata on your errata and say that you were right. "In the months after..." is correct, even if it took more than 12 of them to be so.
> One can easily argue that every dollar spent on interstate highways is earned-back multiple times in tax revenue
One could, but one would be wrong to do so. What you're essentially saying is that all those traffic fatalities -- one 911 per month -- is "worth it." Just the cost of doing business. You know, traffic fatalities have a cost too, in terms of lost productivity, insurance payments, and the like.
You can also make the exact same argument you propose about the return on investment in rail -- that it pays dividends. And it would be a stronger argument. Rail is also safer and more fuel efficient, and doesn't put necessary wear on the subsidized roads.
Hauling freight in trucks across the country is kind of stupid when you could be using trains to do the same thing.
And, putting aside the freight component of the argument -- which is sidestepping the issue of traffic fatalities any way -- it's plain to see that, if we're worried about dependence on foreign oil, we'd do better to use mass transit, and ditch the tar sands oil too.
A passenger train gets about 468 passenger-miles/US gallon. A typical city bus gets about 26 passenger-miles per gallon with just SIX people on board.
A lot of people rag on mass transit because they view it as a subsidy. People don't like the idea of subsidizing something they don't use. Aside from the fact that roads are heavily subsidized too (remember, the federal gas tax would need to be raised 40¢ per gallon to make the interstate system SOLVENT), this omits the fact that urban dwellers subsidize roads to the suburbs that the urban dwellers don't use. It cuts both ways, see?
I think the "excess auto deaths" phenomenon is well established. But I wonder how much of it was really due to the excess inconvenience and annoyance of air security, versus fear of further attacks. (An anecdote: a colleague of mine switched from flying to driving on his monthly long-distance commute at the time, because his terrified wife forbade him to fly.)
If that's the case, then I wonder if we could even attribute some lives *saved* to the reassurance effect of security theater during a very nervous time.
I hate the pointless, intrusive, and demeaning TSA experience as much as anyone, but I recognize that back in 2001-2002 the situation was very emotional and complicated. I flew on September 11, 2002 -- the first anniversary -- and people were still so nervous that most planes were nearly empty and the airline employees were personally thanking everyone for flying that day.
I wonder how much motorcycles contribute to this. They are even more dangerous than cars, and to a certain demographic, nothing says "American freedom" more than riding a motorcycle. Motorcycle registrations have seen a fairly significant rise since the turn of the century, after almost 2 decades of steady decline.
> One can easily argue that every dollar spent on interstate highways is earned-back multiple times in tax revenue
One could, but one would be wrong to do so. What you're essentially saying is that all those traffic fatalities -- one 911 per month -- is "worth it." Just the cost of doing business. You know, traffic fatalities have a cost too, in terms of lost productivity, insurance payments, and the like."
I think it would be right for him to do so. The problems you mention would happen anyway on congested, non-interstate roads. This link supports Les' point that the interstate system was a good investment, esp for productivity. It also makes a safety argument.
39alpha: Heh! You point out that there are two independent reasons that people might not want to fly (terrorists, security theater), and then in your last sentence you sound like you're assuming that all the people who chose not to fly did so because they "were still so nervous" (i.e., about terrorists).
I used to fly pretty frequently, but stopped doing so after the TSA harassed me a couple times. Nervousness about another attack had nothing to do with it.
The general feeling seemed to differ hugely by location. In NYC and DC, you're right, there was a definite anxiety in the air. But there was a very different feeling if you were in San Francisco or Santa Fe.
Like Clive, I'm a bit sceptical about these studies. I would however like to see some figures about the rise of traffic casualties as a result of more widespread cell/smartphone usage since 9/11. I bet they are even more significant than those attributed to people shifting from flying to driving.
Yesterday, I almost ran over a dog whose owner was too busy with his phone to keep an eye on his pet, and on my way to the gym I saw a kid nearly getting hit by a bus while crossing the street texting or updating his Facebook status. It's mind-numbing just to what extent these devices can turn perfectly normal people into complete zombies.
@ Dirk Praet
"Like Clive, I'm a bit sceptical about these studies. I would however like to see some figures about the rise of traffic casualties as a result of more widespread cell/smartphone usage since 9/11."
I'm with you on that. Multitasking has a proven negative effect on both awareness and reaction times. Plus, I've nearly been hit by people messing around on phones and such. So, I'm sure more data on that would be interesting.
The Volt is indeed a luxury car, particularly if you have enough solar panels to not need to buy power for it. Even at prevailing rates for electricity it's cheap to drive (roughly like $1/gal gas).
I happen to love mine. Best car I've ever owned, by a large margin, and I've owned some pretty cool cars.
I know my wallet started to get hard to sit on when I stopped buying gasoline (I'm a cash kinda guy).
I don't make enough money (not even close) to take advantage of the 7.5k tax credit - I only got about 2k of that. The fact that it goes to 7.5k for those with a lot of taxable income is probably not fair, I agree. The fact that the very richest don't tend to show a lot of taxable income is probably a larger factor in the whole fairness equation, however.
In this case, you could consider this a subsidy that actually saves money, in the form of jobs that pay well over at GM, versus letting yet another large company go out of business, and putting a lot more people on the dole - net costs are kind of tricky and have a lot of variables to make certainty difficult.
Let's figure in the costs of all our wars to ensure the flow of cheap oil for starters, and all the deaths they create. The situation isn't simple open and shut.
39beta: Yes, two independent reasons -- but thate first one should be "fear of terrorists," not just "terrorists!"
And I'm by no means claiming that all of the air travel avoidance of 2001-02 was due to fear, but I believe a lot of it was. People were really, really freaked out back then. Today -- with vastly increased hassles and a much reduced (actual and perceived) terrorist threat -- the situation is very different and I expect most air travel avoidance is due to TSA.
People who read Schneier are hardly representative of the general public on security/risk issues, so how you or I make/made these decisions doesn't really tell us much about the wider population. But I'm sure the question could be resolved; I think there were a lot of "are you afraid to fly" surveys taken back then.
There has been a huge push for seatbelt laws and enforcement over the past decade. That makes an enormous difference in traffic fatalities.
Oftentimes travel by plane includes delays. Once, traveling between two cities, the delay was so long I could have driven the distance 3 times while I waited. Service is just shoddy. Folks are treated awfully. Add in the fear of being abused by security, and a few bad first-hand experiences with this joke of security, and driving starts to look far more attractive.
Trains are ludicrously easy to sabotage. The disasters of a derailment often take out entire neighborhoods. Be glad we don't rely on them too much.
When was the last time you took a survey? When was the last time you skipped one? Who do you think has time to take surveys? How badly do you think this skews the results?
Subsidizing the Volt and similar cars encourages adoption in a difficult market. Long-term, this will encourage fueling stations, repair stations, better battery technology, and the rest of the ecosystem we need to break our addiction to oil.
I'm not convinced how much is all airline security. I hate to fly, but not enough to add 16 hours to my travel. OTOH, we do carpool 1000+ miles bc it is cheaper to drive 4 people even with gas prices than buy tickets for them all.
there are other reasons for increases in fatalities too. distractions (mobile phones), carelessness (people are so used to driving and cars have so many safety features people forget it is dangerous).
I bet we'd see a sharp decrease in traffic fatalities if all cars had the driver's air bags and seat belts removed, and required a spike on the steering column aimed at their chest. I say that in jest (I would never seriously recommend it ), but it would make for safer drivers. so much for no texting laws ("the phone can wait, safety first" as he nervously keeps the spike in mind).
--Ok, feel like a bit of an ass; guess I can deal w/ it. But they're not even that good, need a lot of improvements and get timing issues or die altogether too quick.
So where is the trillion dollar War on Bambi?
[sarc]Don't give the Gun Lobby any ideas...[/sarc]
Actually, if your numbers are right, deer are more dangerous to human life (in North America) than wolves, bears, or mountain lions.
Which would be a reason to engage in a War on Bambi. Or at least, a planned culling of deer, to reduce danger to humans.
Bruce, I just wanted to let you know that I had to go the library to get a copy of Practical Cryptography because I couldn't get it for my kindle and waiting to order it over Amazon was just toooo long.
@ Nick P
You're committing a few fallacies.
First, I'm not saying the Interstate system hasn't paid off. At issue is whether expanding rail infrastructure would have been a better investment. The suburbs are not particularly productive: most jobs are in the cities. The interstate connects the suburbs to the cities, though cities can connect by rail just fine.
And second, if fewer people drove because it was easier to take rail, there would be fewer traffic fatalities. Fewer drivers mean fewer opportunities for collisions.
It's also a fallacy to suppose that larger highways mean less congestion. Larger highways make it easier to drive, which leads to more drivers, which leads to more congestion. You can't ease congestion by building bigger highways, you can only ease congestion by giving people transportation options.
Considering the role of deer in the propagation of Lyme disease, their destruction of American life and health is probably much heavier than the numbers you cited.
" At issue is whether expanding rail infrastructure would have been a better investment. The suburbs are not particularly productive: most jobs are in the cities. The interstate connects the suburbs to the cities, though cities can connect by rail just fine.
And second, if fewer people drove because it was easier to take rail, there would be fewer traffic fatalities. Fewer drivers mean fewer opportunities for collisions."
Good points. The trains definitely have advantages in safety, maybe cost too. The main weakness with that proposal is that Americans liked cars, ownership and autonomy. The highway option is superior in those areas. (Privacy too.) Also, how much would the trains or buses help me in rural areas that I often visit? I can drive there along a highway immediately with no waiting, it cost almost nothing, and have my car available at any point.
(Note: There might be a totally equivalent option for people who use alternatives like rails that I don't know about b/c I haven't lived in a place that has many public transportation options. If there is, feel free to tell me.)
As for congestion, maybe and maybe not. More people being on the road as industry expanded was a given. The highways certainly invite more drivers by design, although they're also designed to handle more drivers safely. Yet, there's still plenty of accidents. Then again, the report pointed out that many accidents were prevented as traffic moved to Interstates. So, all I can conclude right now is: who knows. I'd need more data across all use cases before I could be sure which is the overall safer option among those with great national productivity.
> good points. The trains definitely have advantages in safety, maybe cost too
> Privacy too
Not with all the automated police cameras tracking license plates and storing motorist activity in databases.
Privacy and anonymity are different concepts. Pay for a latte in cash, that's not private but it's anonymous. Go to the doctor, that's private, but nowhere near anonymous (the doctor examines your body in minute detail).
>Then again, the report pointed out that many accidents were prevented as traffic moved to Interstates
Accidents and fatalities are different. Ralph Nader and Robert McNamara did a lot to reduce traffic fatalities, though the accident rate may continue to climb (which has a cost).
But aside from the brutal ethical implications of your value judgement, that one 911 per month worth of traffic fatalities is "worth it" for the "productivity" that the Interstate system offers, you're also sidestepping the issue of traffic fatalities by focusing on abstract economic matters.
Maybe you're a communist, and for you the economic analysis is the final analysis, but you're also not really addressing any of my points, and you have yet to contend with the fact that the productivity of the interstate system is heavily subsidized by a gasoline tax that hasn't been raised in 20 years. If drivers bore the full cost of their driving, more of them would seek alternatives. Instead, government subsidies of roads lead to a situation where many people have no choice but to drive.
This doesn't have to be a hypothetical debate: it is straightforward to look to europe where they opted NOT to dismantle their rail infrastructure. Today's abortive efforts in places like Wisconsin to re-establish this infrastructure have to contend with the fact that the rail infrastructure we dismantled was built with government-subsidized land grants in the mid-1800's. Consequently, the cost of re-acquiring the land for that infrastructure is largely prohibitive.
This has furthermore led to problems that are harder to quantify: for example, the suburbs are unsustainable, and depend heavily on the subsidization of the interstate system for their viability. If you look at most cities around the globe, the inner city is where all the wealth is, and the suburbs are the slums. The Interstate artificially inverts this natural tendency, allowing the more affluent to live further away from the real centers of commercial activity in the cities.
In the United States, there are about 9 parking spaces per car. That land has value. Everybody assumes that parking should be a right, but people also want to park in places where that parking space could be reallocated for other more productive purposes.
Cities, then, in effect, subsidize the suburbs, providing jobs for suburbanites who then take their wealth out of the cities, and pay property tax elsewhere, and shop at malls that should be urban retail, all things being equal. Suburbs then turn around and antagonize the cities. The valorization of rural life, and the suburban imitation of this rural life (with all the bourgeois accoutrements of urban life tacked on) led, in the 50's, to the widespread destruction of urban areas.
The interstate was planned before the Civil Rights Act was passed, and it's not by accident that interstates often connect to cities where historical black and immigrant neighborhoods once stood. The demolition of black neighborhoods for interstate construction sparked nationwide riots in the 1960's, against a background of planned neglect and police brutality. So the interstates have a historical cost as well, which is difficult to quantify and underappreciated.
Just as the government's efforts to fight the Global War on Terror are exempt from constitutional constraints, they're exempt from cost-benefit analysis.
It's our patriotic duty to have blind faith and unquestioning trust that each and every measure, classified or otherwise, is necessary to the War Effort and effective at securing the eventual Victory, however that may be defined. The NSA, of course, is continually watching us to ensure that we are not remiss in that duty, which is essential to the War Effort.
I think the analysis fails. Flying is undoubtedly is safer than driving, but US revenue passenger enplanements in 2004 exceeded the number for 2000. While US passenger flights may have dropped in the immediate aftermath of 9/11, there is little evidence that many people avoid flying today because of either a fear of terrorism or the hassle of the TSA.
Quite a thorough reply. I appreciate it. Plenty to think on in the near future. ;)
If those stats are true, those people are LITERALLY being annoyed to death.
That's the literally which means the words used represent reality, not the literally which means "not literally."
A couple weeks ago I was considering how we're not really the land of the free and home of the brave anymore
I don't avoid flying because I'm afraid of terrorism or danger in the air. I avoid flying because I can't stand the security questions/treatment we are put through when going through airports.
I hate it. The risks of death in driving are small enough that I am willing to take them to avoid the nearly certain annoyance and possible humiliation I'll receive at an airport. I think this is a rational choice. And yes, 9/11 is responsible in the sense that airports got worse after the attacks.
I would rather drive 24 hours for my vacation with my family than
1. Pay to park my car
2. Pay for a plane ticket
3. Rent a afar for my stay.
Even at 10 MPG driving is cheaper.
Security as practiced in airports by the TSA is a joke.
Superb overcome! I need to apprentice whilst you modify your internet site, how will i sign up to for the weblog web-site? A accounts made it easier for us a tolerable deal. I have been slightly familiar on this your current over the air given lively apparent notion
Look at the number of total crashes, it didn't change, just as accidents with injuries and property damages only and since the yearly increase in traffic stayed stable you can't conclude that 9/11 had any impact on traffic fatalities.
Schneier.com is a personal website. Opinions expressed are not necessarily those of Co3 Systems, Inc.