Schneier on Security
A blog covering security and security technology.
« Turning off Fire Hydrants in the Name of Terrorism |
| The Doghouse: Tornado Plus Encrypted USB Drive »
September 12, 2008
Cost/Benefit of Terrorism Security
"The terrifying cost of feeling safer," from the Sydney Morning Herald:
Sandler and his colleagues conducted an analysis of the costs and benefits of five different approaches to combating terrorism. I must warn you that, because of the dearth of information, this study is even more reliant on assumptions than usual. Even so, in three cases the cost of the action so far exceeds the benefits that doubts about the reliability of the estimates recede.
Because the loss of life is so low, they measure the benefits of successful counter-terrorism measures in terms of loss of gross domestic product avoided. Trouble is, terrorism does little to disrupt economic growth, as even September 11 demonstrated.
Using the case of the US, Sandler estimates that simply continuing the present measures involves costs exceeding benefits by a factor of at least 10. Adopting additional defensive measures (such as stepping up security at valuable targets) would, at best, entail costs 3.5 times the benefits. Taking more pro-active measures (such as invading Afghanistan) would have costs at least eight times the benefits.
According to Sandler, only greater international co-operation, or adopting more sensitive foreign policies to project a more positive image abroad, could produce benefits greater than their (minimal) costs.
What's that? You don't care what it costs because no one can put a value on saving a human life? Heard of opportunity cost? Taxpayers' money we waste on excessive counter-terrorism measures is money we can't spend reducing the gap between white and indigenous health -- or, if that doesn't appeal, on buying Olympic medals.
Posted on September 12, 2008 at 6:32 AM
• 34 Comments
To receive these entries once a month by e-mail, sign up for the Crypto-Gram Newsletter.
"security" is made up of three things, first, propaganda to keep the fear high. Second, blatantly eliminating the bill of rights and making any crime legal with a badge.
Third, brute force and stupidity against the innocent population whom the "national security state" deeply distrusts.
In the '60's and 70's national security states were established in many south american countries. This was done by graduates of the school of the americas, (school of coups)
The result was some negligible formation of resistence groups who were largely ineffective and small, serving only to help the junta's sell the fear, and total distrust of government that reasonably continues to this day as it is the natural response of the people to a corrupt system. Totalitarian systems are always corrupt, absolute power corrupts absolutly.
At least one quarter of the population will always support a system like this, (google 'fascist personality)
Bruce,it's really your work, to write blog? It's important, but I don't believe in it :)
Ross Gittins is the economics editor for the Sydney Morning Herald. He's like the Bruce Schneier of economics. He brings a lot of common sense and insight to the table, he relates it to the wider world, and stuff he says is so sensible, it's startling.
Read his other columns. One of the more interesting things he has in common with Bruce is this: he relates psychology and the way we think into economics (or in this case, security).
Is it just me, or does anyone else think that September 11 is not a good data point supporting a premise that terrorism does little to disrupt economic growth?
Personally, I think a better way to think about these cost/benefit scenarios is to compare predicted-lives-saved among different interventions.
What's the PLS from preventing another 9/11-style attack? From preventing a shoe bombing? What is the cost-to-PLS ratio there? What about the cost-to-PLS ratio of (say) better highway traffic enforcement, or some other intervention that would cut down on traffic fatalities? What about the cost-to-PLS of providing affordable healthcare to all? Improving schools?
In general, I think we ought to be able to think about the rough cost it takes to prevent a death due to various causes (I'm SURE that insurance companies do this all the time). From what I understand of the general setup, all the cost we've put in to preventing terrorist attacks are way out of proportion to the number of lives saved.
The stat's for terrorism are interesting only to the point when you become one.
To die because x wants to avenge his gripe against y --is unpalatable to most.
Fear like love doesn't yield to economic analysis.
It seems to me that the only way to get a sensible terrorism policy is to find a way to terrify the public about the opportunity cost of counter-terrorism measures. Something like "1 million children will die over the next 20 years because X was spent on the Iraq war instead of on Y, which could have saved their lives. One of those children could be yours." Do whatever it takes to make the fear of misplaced spending as real and tangible as terrorism.
"In this we're [Australians] much more accommodating than the Americans. In the Land of the Free, it seems, people are a lot less willing to give up their freedoms."
This came to mind, for some reason -- from Thomas Paine, following a different September 11 event:
"Those who expect to reap the blessings of freedom, must, like men, undergo the fatigues of supporting it. The event of yesterday was one of those kind of alarms which is just sufficient to rouse us to duty, without being of consequence enough to depress our fortitude. It is not a field of a few acres of ground, but a cause, that we are defending, and whether we defeat the enemy in one battle, or by degrees, the consequences will be the same."
(Of course, the LotF has since made nice with the U.K.)
I'm not sure if economic analyses will sway those giving up freedoms in their fight to protect Freedom. If they run out of money, they'll just print more.
Excuse me, sooth sayer, but if fear "doesn't yield to economic analysis", then as a social issue it yields to no analysis at all. And yet, almost our entire national politics is driven by fear.
The antidote for fear is reason. This sort of analysis is in fact _precisely_ what would cure our politics of its prevailing unreasoning fear, were it only possible to educate people to apply their reason to risk.
If we fail to ascribe a finite benefit to a security measure, then we are in effect saying that the benefit of such a measure is infinite, dwarfing any cost, whether economic or political. That's the road we've trodden towards our burgeoning high-surveillance police state.
It is perfectly appropriate to point out that it is irrational to fear (for example) nuclear annihilation -- which has killed 220,000 people in the 58 years since 1945 -- while embracing life as an automotive commuter, when auto accidents kill 50,000 people every year -- that is, a Hiroshima/Nagasaki bombing campaign every 4.5 years.
If the question is specifically WMD terrorism, it is also perfectly legitimate to ask the question "How often should we expect to lose a city to WMD?" The answer cannot possibly be "never", because it probably wouldn't even be that for North Korea. A rational analysis would set that number to "Once a century", or "once every 300 years", and then calibrate security resource allocation to achieve that kind of threshold.
For comparison, to help think about what that number should be, it is worth recalling that we lost New Orleans three years ago (without the kind of hysterical political-economic overreaction that would have resulted had it been nuked instead of blasted by Katrina), we lost San Francisco in 1905, and we lost Chicago in 1871. None of these events brought about the end of the world, or even the collapse of the U.S., or even a noticeable amount of economic discomfort. That suggests that perhaps the rational threshold is somewhat less than one major city lost per century (so as to be less than the background level of loss).
During the cold war (when the risk of annihilation was much higher), we used to know how to make these sort of scary trade-offs. Apparently we've lost the knack now, though.
If we refuse to at least attempt to consider the risk rationally, we will make all our decisions out of fear. When we demand 100% assurance from our government that we will be safe, we are asking the impossible. When we ask that they take every possible measure to combat terrorism, irrespective of cost, we are inviting appalling consequences, not least of which the destruction of our liberty.
Irrational fear is a magnificent motivator - leadership of all colors have always found a use for it. Cultivate it, and you might be able to bring mobs of disparate groups together, but like much in the way of irrational motivation, it's unstable. Beaten dogs will eventually bite their masters.
There's a distinct difference between fighting terror and fighting terrorism/terrorists. One is about the fighting the fear, while one is about addressing bad-actors and their terror-inducing actions - actuarial stats aside, question is: which fight are our leaders fighting? How is the media skewing and spinning? Is willful promotion of irrational fear itself a form of terrorism? Can we actually expect that the general masses are smart enough to interpret even the most brilliant statistical analysis while in panic-mode?
Fear is a killer to rational or critical thinking. The article (and heck, this blog!) is part of what helps build perspective/understand the dynamic of fear - I hope someday it'll help frame some real progress.
BTW, my kid is afraid of the dark....could we rally all the other kids similarly afflicted to help fund that solar space reflector concept so we get perpetual sunlight to fight this terror? I told him he was being irrational, but you just try to reason with a 3 yr old.
You may be willing to accept 1 hiroshima a century -- I am not . others might be willing to trade 10 .. of course we should adjust for inflation in population.
You can keep indulging in this analysis and the result will always be incomplete because the premise can't be established.
You are wrong that national policy is driven by fear -- may be greed.
Reason is not as strong as fear which seems to be wired into every living being for self preservation -- you are not behaving any differently than any other; as 9/11 goes further into the past the fear recedes too.
Reason should tell you that time actually is working against you now (your 1 in a century calc) -- but you are arguing against it!!
It's not just you. Economic analysis is an imprecise science, I suppose, but it stretches credulity to claim 9/11 had no significant economic downside.
Is "terrorism," fraught as it is with political implications, really amenable to economic cost-benefit analysis?
That said, I do think it is time that We the People start asking legitimate questions about whether the "security" measures our Leaders are forcing on us are effective, justified, and provide value for their costs (measured in intangible freedom and liberty along with tangible dollars). The bushies have been spectacularly successful in stifling any such questions with massive doses of fear and branding both the questions and the questioners as "unpatriotic." Now that the bushies are on their way to billionaire-Emeritus positions at Halliburton and Blackwater, it is time to ask the questions and have the debate that our current Leaders have done everything possible to prevent.
If we look at the direct costs of 9/11, ~3k killed and the estimates I've heard say around 15-30 billion (call it 3e10) dollars of direct and indirect costs. Compare that with the Iraq war, supposedly to prevent the "next 9/11" or some such nonsense. It has cost somewhere between 800 billion and 1 trillion (1e12) dollars and lead to the deaths of over 4k us troops, and probably 100,000 civilians in Iraq. That's several orders of magnitude higher.
As for the economic impact of 9/11, I doubt it was a large as some say. I was laid off 4 months prior--when the first dot-com era imploded. The evaporation of the inflated stock market had a lot more to do with it than the twin towers falling.
It was a terrible day, but let's keep some perspective.
Professor Sandler may be ignoring game theory in suggesting the US "adopt more sensitive foreign policies."
While "sensitive" is vague, the clear implication is that the US should prefer policies more likely to meet with the approval of terrorists and would-be terrorists.
Perhaps -- perhaps -- we could appease Al Qaeda in that fashion, but then, of course, the next group that wishes to affect US foreign policy would know exactly how to do so.
I'll comment on the author's conclusion, that only positive image building can be efficient. In 1990, an expedition to Nanga Parbat in northern Pakistan brought along ace climber, Everest summitteer and medical doctor Morten Rostrup. Dr. Rostrup never climbed Nanga Parbat, because rumours spread that a doctor with medicine was in basecamp. The life expectancy in this area is around 52 years, and people travelled for days at high altitude to get there, including people all the way from Afghanistan.
Dr. Rostrup treated and saved many lives with his own two hands and very limited resources, but he made a huge difference to the people he helped.
I think that if the attempt was made, it would be a lot more efficient way to combat terrorism to save lives rather than take them away.
Whoa... "terrorist-approved" stamps on foreign policy initiatives sounds overly cynical. We're fighting a "war" against some really angry folks who've (rightly or wrongly) singled out targets as symbols of their cause. They're likely not part of the political or conventional life fabrics of most societies, so I doubt that there are enough angles a foreign policy/administration can take to make this tenable.
We're likely not going to have direct foreign policy impacts on terrorists, but rather indirect. This will mean that we'll truly have to understand the root causes of this kind of hatred....but when we're fighting a symbolic war based on dogmatic, faith-based ideals, this is a very difficult cultural challenge that we've been notoriously bad/clumsy at addressing to date.
@ sooth sayer: yes, deaths from terrorism are just a statistic until you (or a family member) become one. So are deaths from being hit by a drunk driver, and deaths from antibiotic-resistant infections, and deaths from mad-cow disease. Is dying from one of these ways more palatable to you than dying from terrorism?
@ False Analogy: It's true that some accident deaths are partly attributable to actions by the dead. So let's avoid that by looking at people killed by drunk drivers (excluding the drunk driver and even the passengers in the drunk driver's car): there were about 2,500 of those in 2006, according to the NHTSA. Closer perhaps to manslaughter than murder, but pretty comparable to terrorism deaths as far as innocent victims goes, right? And over any decade you can name, an order of magnitude larger than terrorism deaths.
The point, False Analogy, is not what we fear, but how we deal with risk. Simply put, spending hundreds of billions of dollars to "fight terrorism", while failing to invest seriously in New Orleans levees is an objectively stupid decision, from the point of view of resource allocation. The source of that stupidity is fear.
I would have thought that a clear thinker would understand a simple point like that.
"Is it just me, or does anyone else think that September 11 is not a good data point supporting a premise that terrorism does little to disrupt economic growth?"
Great point. It's not a non-informative prior, but its sole use cannot create a state of knowledge. Israel, India/Pakistan, and much of Africa would be useful in creating a much larger data set on economic impact (provided, of course, that you could develop a useful metric or series of metrics that would explain what economic growth actually *is*).
"is an objectively stupid decision"
Really! And what makes that decision "objective"?
@petej, et al
I believe this whole economic discussion is muddied by the fact that many if not most western political leaders point to Sept 11 as the start of the economic woes we are currently experiencing...and this is certainly supported by the fact that in the months following Sept 11th, the US economy was clearly stalling, with airline-related industries in particular struggling to stay afloat. Few remember that in the months PRECEDING Sept 11th, the US economy was clearly stalling, with airline-related industries in particular struggling to stay afloat.
Whether or not there was a significant impact can be debated, and I'm fairly certain there was. But I think it's also worth noting that it's been tagged as the root of all of our problems as a cheap political blame-avoidance mechanism for years. The US economy has been stagnating since May 2001, at least.
Security is the opposite of what most assume. You actually feel secure when there is less to scare you, which means fewer barriers, no checkpoints...open and unfettered spaces with nothing to worry you. The controls are invisible, and freedom is the ultimate security. The moment a control falls in front of you, like a useless TSA checkpoint, you will feel less secure because you are now less secure.
That is retarded, 9/11 was NOT the start of the economic woes faced by the United States, the trillions of dollars wasted on the war in Iraq have had at least 1000x an impact than 9/11 did (and probably at least 20-100x than Katrina). Since 9/11, there has also developed quite a culture of fear in the U.S. which permits huge pork-barrel and security-theatre expenditures (the TSA and the DHS, for example) which have probably cost a lot more than the immediate economic costs of 9/11 did.
Its so frustrating to watch the United States slowly self-destruct (or slip towards fascism, take your pick). As recently as 10 or 15 years ago, you guys had a great country, a country that most of nations of the world respected, and some of them even liked... The Bush administration has almost single-handedly ruined that. I seriously think historians are going to look back on this decade in U.S. history and compare it to the fall of Rome. Like the U.S.S.R. in the 90's, the U.S. is now a decaying shell of an empire. There's still strength there, but your nation really seems like its rotting from the inside out. You have right-wing crazies running your government and controlling most of your media. And most of you don't seem to care about it. The fact that McCain even has a chance in this election, says a lot about the gullibility and stupidity of the average American. Damn, its so frustrating to watch a formerly great nation wither and die like this.
well the warwas waste of time and money infact this war destrroyed the pestige of ameica amongst the muslim countries who once loved america as much they love their counties only a few small group of people degraced islam and entie muslim wold andamerica instead of showingit's intelligence just started war with otther muslim nation suspected of been terrorist.america should revise it's policy reguarding terrorims this would help to restoe it's image and whole world ecocomy especially central asia
and yet, people are trying to save $10 when they buy cheaper tires for their cars!
and they try to save 30 seconds by speeding thru a "dark orange" traffic light!
seems to me the risk of being killed in a traffic accident is a LOT higher than in a terrorist attack
One other thing to remember is that these "security measures" may well be disruptive to "economic growth". Thus their actual costs may be greater than just implimentation and operation.
At the same time they could do nothing to improve actual security (possibly even reduce it.) Especially when the people making the decisions are publicity hungry politicians. (Whereas an actual security expert may have ideas which are counter intuitive, "against common sense" or to "do nothing". The latter being against a politician's nature.)
It's even possible that expensive "security theater" will not even make the average person "feed safe". It could just as easily make them feel more threatened or arouse other emotions such as anger. As well as creating new risks. e.g. planes where one (or both) of the pilots have been wound up by some "screener" before they even get into the cockpit. How do you ensure that the people carrying out the "security" are not thieves, smugglers or terrorists? Etc.
There was an economic study done a while back where a bunch of economists tried to figure out what value society places on a human life, precisely by looking at cost/benefit decisions taken involving people's lives.
Think they figured it out to be about $6 million. I'll try and dig out the title and its authors...
@Carlo: You're right about the kind of analysis that should be conducted. But running a country that way is inconsistent with democracy because fear shuts down the rational mind.
During every war the US fought up to about 1960, including the Cold War, the way "the system worked" was not democracy. The government censored the news -- and publicly admitted that they were doing it not just to keep secrets out of enemy hands, but also to maintain morale among civilians -- and most everyone accepted it as a wartime necessity.
But in Vietnam and subsequent wars, improved communication and transport tech has given newsies the ability to evade such restrictions most of the time, allowing the masses to see what is going on and successfully demand stupid action.
Perhaps our system needs to be modified to work more like it used to. Or perhaps the genie is out of the bottle and can't possibly be put back in. Either way, so far as democracy is relevant, it is the problem, not the solution.
As a small scale example of the cost/benefit mismatch, consider the costs of the US No-Fly list for this Canadian "businessman whose name is one of the many that have erroneously landed on the U.S. Department of Homeland Security's flight passenger watch list [and who] has decided to change his name to avoid lengthy security hassles at the airport" as described in this article:
What about the costs for US Homeland Security, which apparently wastes 2 to 6 hours detaining and questioning this fellow every time he flies?
And the benefit is what, precisely? Fortunately, no real terrorist would ever think to change his name.
Bruce, you've gone into the problem with high-value/low-frequency attacks, but there are also issues with low-value/high-frequency attacks. If a $100 asset must be placed where it could be destroyed 10 times in a year, $500/yr to protect it is actually a bargain. But to anyone who doesn't understand the math, $500 to protect $100 looks foolhardy.
"While "sensitive" is vague, the clear implication is that the US should prefer policies more likely to meet with the approval of terrorists and would-be terrorists."
That's not the intended meaning of a suggesting friendly foreign practices. if USA maintains lot of friends, terrorists will find few disgruntled supporters (in eg, Canada, Berkeley, Hollywood, or France :-) ). there are actually countries in the Mideast that do not want resident al qaeda (one good reason: those countries would be "playing with fire", much as Republicans are discovering that *their* useful idiots come with an (only political) cost.)
It is difficult to judge the benefits from terrorism security, since you can not measure it directly. By how much percentage was reduced the chance for terrorists attack? Nobody can tell that. Therefore expenditure on terrorism safety can not be easily controlled and justified. This give a free hands to government to spend as much money as they consider it to be necessary into security, without justifying it.
Schneier.com is a personal website. Opinions expressed are not necessarily those of BT.