Krugman, Mueller, and McCain on Terrorism

Mueller’s argument (which I read when it appeared in Regulation) relies on the (implicit) assumption that terrorism and “lightning, accident-causing deer, or severe allergic reaction to peanuts” are mutually exclusive. But they are not. Terrorism is an additional threat to life and limb.

If terrorism never rises beyond what it is today (one Paris like attack every few years or so), then we can rest easy. However, the Muslim population is growing in the West, and there seems to be no indication that radical Islam as a force is abating.

Let’s say that there are 5 million Muslims in France. Imagine we get to the point where each year .01 percent of them decide that they want to die for Allah (not unreasonable at all, I’d bet). That’s 500 terrorists a year.

In the Paris attack, 8 terrorists killed about 150 people. That’s about 19 deaths per terrorist.

Thus, if we have 500 terrorists a year killing 19 people each, that’s 9,500 deaths a year. A huge problem. Even 100 deaths a month would paralyze the economy and likely set off a mass exodus.

You can’t just wish away the overreaction, it’s part of human nature.

In conclusion, Western leaders are playing with fire through their immigration policy.

I understand the argument, but I only agree with it up to a point. We still exercise precaution to minimize the risks that we are frequently exposed to. I am more likely to get in a car accident than killed by a terrorist, but that doesn’t mean I am going to just buy any lemon. The challenge is deciding how much we are willing to spend to minimize the risk of a terrorist attack, not whether we should minimizing it at all. I don’t think you would necessarily disagree with that, but some libertarians really don’t seem to take the threat of terrorism seriously.

Deer and peanut allergies are pretty dangerous, though, and there are measures taken by gov’ts to prevent agäinst thēm.

Never mind… (I’m going to go hide in the corner with the dunce cap).

A point I thought of the other day when someone I was discussing this with raised the question: “why just target airplanes? Why wouldn’t they just blow themselves up in the airport lobby in a big crowd of people? They’d get just as many.” One advantage to targeting planes is that it’s just about possible to enact severe restrictions to boarding a plane, enough to make everyone’s life a little bit more miserable. If they targeted just big crowds, that’s not something any feasible law can address. So possibly a terrorist gets more mileage out of attacking in ways that it’s plausible governments will respond to – leaving a much larger legacy than they would if all they did was kill some people and make everyone else feel uneasy for a little while.

@David Henderson
Not quite. Krugman (and you?) seem to think that the only ‘wrongheaded response’ is to (over)
react via some allegedly counterproductive Bush-Cheney counterattack. Another ‘wrongheaded response,’ but alas an all to typical one, is for the one(s) attacked to turn inward and blame themselves for the attack, lose hope, and eventually surrender. Especially if the attacks are multiple over a long period of time, and aimed at wearing down the victim. (“Terror is not a tactic, it is a strategy.” TM) Have you ever known a woman in an abusive marriage? Alas, my daughter has been in two of them. Repeated attack, wear down the victim, victim loses confidence in herself, eventually adopts attacker’s view of herself… Fortunately, my daughter woke up after a while, but will Western Civilization? Or rather will our Dear Leaders wake up? They, alas, seem deaf, dumb, and blind to the threat…not to mention hopelessly ignorant of Islamic and Arab history.

the number of Americans killed over the same period by lightning, accident-causing deer, or severe allergic reaction to peanuts

Aren’t you ignoring the non-zero probability of a catastrophically destructive attack such as a dirty bomb (and the resultant psychological and economic damage)?

Even if the probability of such an attack is low, its “expected value” is great enough to warrant increased caution in Europe, including a re-evaluation of Merkel’s open-door policy.

[Comment removed pending confirmation of email address. Email the [email protected] to request restoring this comment. A valid email address is required to post comments on EconLog and EconTalk.–Econlib Ed.]

@John T. Kennedy,
Am I wrong in assuming McCain is a fan of the TSA?
I don’t think you’re wrong. Weird, isn’t it?
@Raghav Aras,
In other words, the dead bodies are not the greatest harm, but the response we conceive is.
You’re writing as if you think that government’s responses have not caused many more people to die. But that’s not true. I think you and I can agree that TSA was a response to 9/11. Given the cost that TSA has imposed on air travel, and assuming a reasonable elasticity of demand for short flights, when we add the assumption that even 10% of short flights will instead be made by car, TSA has now caused more people to die than were murdered by the terrorists on 9/11.

I would not so quickly dismiss the idea that far fewer people would fly without an invasive TSA screening process. The short term flight/driving tradeoff likely pales in comparison to the overall reduction in flyers from returning to a pre-9/11 TSA. Under those assumptions, TSA actually saves lives. You may call a decision to fly without an invasive TSA irrational, it’s another thing to say it wouldn’t happen in the real world. Our collective “irrational” reaction terrorism still represents real costs in the world above and beyond the discrete event. Arguing that fear shouldn’t exist is conceptually distinct from arguing that it doesn’t exist and impose real costs on our economy.

Government’s duty is to protect one person from harm knowingly caused by another person, not the choices a person makes himself.

But if the government’s actions cause individuals to choose relatively more risky options (driving) vs relatively less risky ones (flying), isn’t it failing in that duty? After all, my death in a car accident can be the result of other people’s decisions (getting behind the wheel of a car drunk, for example).

@Jon Murphy

Well, this issue is not one of the number of deaths, as understood by Krugman. It is of government’s measures against one particular source of death: homicide.

(otherwise, one can discuss any measure the govt takes that influences people’s choices: flights vs car travel is but one example).

Suppose the French government does absolutely nothing in response to this carnage. Suppose another band of terrorists repeats this in 6 months in Paris, another 150 killed. Can the French President then say: “we estimated that any measure we could take would lead eventually to more deaths since they would influence people’s choices, so we decided to do nothing”? I wouldn’t think so.

The harm an individual brings upon himself is different from a harm one individual knowingly does to another.

My opinion about Islamic extremism is different: I think it is mainly a result of Islam’s teachings. We don’t see Vietnamese extremism, or Korean extremism exported to other countries. So personally I don’t think US response to 9/11 had a major impact on Islamic extremism.

As early as five days after 9/11, Krugman effectively blamed the attacks on a lack of TSA. Since all we need to do to prevent attacks is federalized airport security, we really do have little to fear – we’ve taken care of the problem. Indeed, he insinuated (nearly asserted) that the attacks would not have been possible with government-run airport security.

Apparently the security staff at concert venues in France need to be trained by and provided by the government as well.

So much smoke without noticing the fire!

Whether or not Henderson, Krugman, et. al. are correct, there is a problem which is mentioned, but ignored by these people – human nature. These people are saying human nature is flawed because statistics show that human responses to various threats are irrational. No kidding! It is irrational to assume that he irrationality of humans can be ignored.

The real world demonstrates the human response to terrorism leads to a diminution of liberty via several routes. That is the primary damage terrorism inflicts. Though the topic is ignored by Henderson, Krugman, et. al., their prescription to dealing with the terrorism’s primary problem is to try to instill rational thinking on an irrational populous – a populous which not listening.

The remedy for the danger of terrorism prescribed by Henderson, Krugman, et. al. will lead to less liberty.

The analogy to lightning is a telling one. “Struck by lightning” is usually almost always meant in the sense of something that is truly, utterly, completely random. The classic example of the extremely rare, extremely negative tail risk that one just has to accept as the price of living.

Except that it’s not. “Struck by lightning” is only fully random to an actuary. To any normal person with actual everyday decisions to make, there is an obvious cue that affects the probability of getting struck by lightning: Storms. When the sky is clear, the risk of being killed by lighting is zero. When you see a flashing thunderhead 20 minutes away, you become distinctly aware that risk, while still extremely small, is also definitely not zero.

Most importantly, we do use that knowledge to inform our behavior. If there’s lightning in the distance, or we think a thunderstorm is imminent, we alter our plans. Most obviously, we suspend sports events. Doing so is explicitly written into the rules (http://www.littleleague.org/learn/programs/asap/lightning.htm http://www.littleleague.org/Assets/forms_pubs/asap/WeatherItsSafeToPlay.pdf). I am sure it is possible to come up with a model in which this decision is sub-optimal, but if that were done all it would do is prove that the model is stupid. When 7-year olds whine to their parents about being kept out of open fields during a thunderstorm, it’s not the parents behaving irrationally. The rate of death due to lightning strikes has a selection bias built in. It is partially dependent on the fact that people alter their behavior in response to (broadly accurate, if imprecise and subject to bias) changes in perception of the risks of getting struck.

Having clarified what actually happens with lightning, what does this mean for the terrorism analogy? Is terrorism actually like lightning? If so, then it would be rational (in the economic sense of behaving consistently) to respond to it like lightning- go through most of life knowing that risk is zero, and intentionally alter our behavior in the circumstances we know it is uncomfortably larger than zero.

This is where the lightning analogy gets ridiculous, and Mueller’s argument goes bad. No one takes seriously the guy who says, “If you don’t send the kids out into the thunderstorm, then Zeus has won!” We don’t lionize people who, as a matter of principle, intentionally expose themselves to lighting risk and get themselves struck; we call them idiots who (we think but don’t say because its impolite) asked for it. If (more accurately, “when”) we respond to terrorism in this way, we do take seriously the “…then the terrorists have won” argument. And yet, this is exactly what Mueller recommends, despite his earlier proposition that we not worry about terrorism because over 40 years the risk of death to Americans is on the same order of magnitude as lightning.

If death-by-terrorism and death-by-lightning-strike have similar statistical profiles, and yet we recommend starkly different strategies to deal with that risk, what does that say? It says that we all understand that there are important differences in the nature of the risk, and that we know that the solution involves group behavior.

In econ/game-theory-speak, what terrorist are doing is trying to create a repeated-game prisoner’s-dilemma problem. If individuals cannot coordinate, then the rational response is as described above- avoid the situations in which the risk of death-by-terrorism gets uncomfortably large. That is what “terrorists win” looks like. Now, as any economist would know, the key in a repeated-game prisoner’s dilemma problem is coordination. If every individual can be convinced that the coordination strategy is superior to the everyone-fend-for-themselves strategy, then it can stick and will indeed actually be superior. That is what “we don’t let the terrorists win” looks like. Organizing and promoting such a coordination strategy is exactly what political and intellectual leaders are supposed to do, so in that sense Obama, Krugman, and McCain are doing what they ought.

However, also as any economist knows, coordination strategies in game-theory settings are extremely fragile and borderline impossible to recreate once they fail. They are particularly difficult when coordination must take place in the absence of any explicit enforcement mechanism. “Enforcement mechanism” in the context of this terrorism game theory problem entails things that are all fundamentally incompatible with a free society. Thus, there the coordination strategy must be based on trust and goodwill.

Furthermore, any economist should also know that as soon as any randomness is admitted into the model, the odds that a coordination strategy succeeds plummets when the repeated-game is open-ended rather than closed-ended. Here is where the Obama/Krugman-line of thought is not only wrong but deeply irresponsible given their positions. Anyone organizing a coordination strategy must as part of that strategy organize it from a closed-ended perspective. In ordinary English that means treating terrorism as a problem to be “defeated” rather than “managed”. As soon as the organizer admits that the problem is open-ended, then failure becomes vastly more likely. Thus, the organizer must describe the problem as closed-ended even if he privately disagrees. Because of the “trust and goodwill” that is also necessary for the strategy to work, the organizers not only have to say the right things but also act in ways that are consistent with them in order to maintain credibility. If you aren’t capable of doing that, then you have no business being a leader.

“dignify it with the name of war”

That phrase really rubs me the wrong way. War may sometimes be a necessary activity but it is never a dignified one.