Charles Wheelan, Ph.D. The Naked Economist

I settled into my seat for a flight a few days ago from Chicago to Washington, D.C., and opened my book: "What Makes a Terrorist." I now recognize that this probably wasn't the best choice of reading material for a crowded plane.

But it is a good read. The author, Princeton economist Alan Krueger, has written an accessible and interesting book on the causes of terrorism. (Disclosure: Krueger was my statistics professor in graduate school; I did badly in the class, which is why I now happily write articles like this that involve no math.)

Number-Crunching the Threats

The full title of Krueger's book is "What Makes a Terrorist: Economics and the Roots of Terrorism." His work (here and elsewhere) is a nice example of how economics can be used to examine issues far beyond interest rates and stock prices. After all, the point of economics is to explain why individuals and organizations do what they do, including people who blow things up.

More important, the findings outlined in the book provide some intellectual traction in the battle against terrorism. Krueger draws conclusions based on an array of data gathered on terrorist attacks around the world, including information on the terrorists, their country of origin, and the country targeted in the attack.

Experts need to study such data for the same reason that epidemiologists study public health data -- the findings often suggest a strategy for confronting a threat, whether it's heart attacks or terrorist attacks. Statistical analysis allows researchers to isolate the effects of certain factors, such as smoking in a public health context or religion as it might relate to terrorism. (I did well enough in Krueger's class to appreciate the importance of this kind of statistical analysis, if not necessarily well enough to do it competently.)

So what makes a terrorist? Here are Krueger's big findings.

Wealth, Education, and Terror

1. Terrorists are neither desperately poor, nor poorly educated.

Instead, "terrorists tend to be drawn from well-educated, middle-class or high-income families." (The 9/11 Commission came to the same conclusion.) But why?

Krueger hypothesizes that since terrorists are motivated by political goals, the more educated and affluent have a greater stake in changing outcomes. He draws a parallel to voting: "Having a high opportunity cost of time -- resulting, say, from a high-paying job and a good education -- should discourage people from voting, yet it is precisely those with a high opportunity cost of time who tend to vote. Why? Because they care about influencing the outcome and consider themselves sufficiently well-informed to want to express their opinions."

I would offer a complementary explanation, which is that affluent and educated citizens are more likely to be rankled by political repression, which appears to be a strong causal factor for terrorism. The citizens with the most stake in a society -- the educated and the affluent -- are bothered most when a newspaper is banned or opposition politicians are jailed. Who would take greatest offense in the United States if the government banned Fox News or NPR? Not the homeless.

Krueger, ever the labor economist, also points out a somewhat macabre explanation for well-educated terrorists: Terror organizations accept the most capable volunteers -- just like any other organization that cares about its success. The bunglers get turned away.

No Political Safety Valve

2. Political repression in a country is consistently associated with a higher level of terrorist activity.

This includes the suppression of freedom of expression, freedom to assemble, and other civil liberties and political rights. This should be fairly intuitive; when these rights are curtailed, the steam has no way to escape the pot. Krueger writes, "When nonviolent means of protest are curtailed, malcontents appear to be more likely to turn to terrorist tactics."

To my mind, this finding presents the greatest dilemma for U.S. policymakers. Some of the governments that have been most helpful to the United States in terms of fighting extremist groups, namely Pakistan and Saudi Arabia, have also ruthlessly cracked down on domestic political freedoms. Thus, these governments are simultaneously fighting terrorists while breeding new ones.

Think Globally, Terrorize Locally

3. Like politics, most terrorist attacks are local.

The Sept. 11, 2001, attacks are an anomaly in this respect. The role that proximity plays can be observed based on the national origin of the foreign insurgents captured in Iraq. When Krueger controls for other variables related to the home countries of the foreign insurgents -- such as GDP and predominant religion -- being from the neighborhood turns out to matter a lot.

There are a host of possible explanations for this. Citizens from neighboring countries may have a greater stake in the Iraqi outcome. Or it may be easier for them to blend in. Or it may simply be easier (and cheaper) for them to get there. Whatever the reason, the data show that "distance seems to be a significant barrier to terrorism."

How Region and Religion Figure

4. Countries that are occupying all or part of another country are more likely to be subject to a terrorist attack. And countries that are occupied are more likely to be the origin of an attack.

Again, this is fairly intuitive, but it has significant implications for foreign policy, in the United States and elsewhere. It strikes me that there's a parallel with domestic repression: If we send forces somewhere to crack down on terrorists, we may be simultaneously creating the circumstances that promote such terrorism in the first place.

5. It's not a Muslim thing.

Countries with a high proportion of Muslims aren't significantly more likely to be the origin of terrorist attacks than countries with a high percentage of some other religion, once civil liberties are taken into account. In fact, international terrorism was less likely to occur between pairs of countries with different predominant religious groups.

Krueger writes, "My interpretation of these results is that religious differences are among the many potential sources of the grievances that lead to terrorism. They are not the only reason for such grievances, and such grievances are not specific to any one religion. Although the world's attention is currently focused on Islamic terrorist organizations, they are by no means the only source of terrorism."

Damage Done

6. The direct impacts of terrorist attacks are not huge.

This isn't meant to diminish the suffering of those directly affected by 9/11 or any other such attack; it's meant to put the scope of the attacks in perspective. The number of people killed in car accidents in 2001 was 10 times the number of people killed by terrorists (my observation, not Krueger's).

Krueger does make the case that economic damage caused by 9/11 was modest given the size and diversity of the U.S. economy. Even the psychological impact dissipated relatively quickly.

His point is that terrorism can only succeed by sowing fear and overreaction, not by destroying things and killing people. He posits that terrorists strike democratic countries more than autocratic regimes because public reaction matters more in open societies. For example, he points to evidence that terrorism has affected electoral outcomes in Spain, Israel, and the United States.

"What Makes a Terrorist" includes a wonderful chart showing the relative risk of dying from assorted causes. The lifetime risk of dying in a motor vehicle accident for the U.S. population is 1 in 88. The lifetime risk for dying of suicide is 1 in 120. The lifetime risk for dying in a terrorist attack is 1 in 69,000, which is significantly less than the risk of dying from a lightning strike. (To be fair, the average American is far more likely to die at the hands of a terrorist than from a shark attack; that lifetime risk is 1 in 3,700,000.)

The point is that terrorist attacks succeed because of the terror, not the attack. We should respond with that in mind.

The Most Reliable Weapon

Krueger's work and other studies like it have obvious limitations, beginning with the challenge of defining terrorism. The data are inherently hard to collect; the interpretation leaves a lot of room for ambiguity and interpretation. For example, should an attack on a McDonald's in Pakistan be treated as an attack on Pakistan or on the United States?

Still, the approach is spot on. Fighting terrorism begins by understanding it.