Terrorism and Poverty Posner Comment

A former president of the World Trade Organization, the current British Chancellor of the Exchequer, the House of Bishops of the Episcopal Church, and many others have claimed without citing any evidence a close connection between poverty and terrorism. Poverty may be related to terrorism, but in ways that are far more indirect and weaker than alleged.

Any sizeable effect of poverty on terrorism is not apparent from what is generally known about terrorist activities. The suicide bombers in the 9/11 attack were mainly highly educated Saudis, not poor Moslems from other parts of the Middle East, Asia, or Africa. The Basque region of Spain may not have done well economically in recent decades, but the members of its ETA terrorist organization are generally middle class and reasonably well educated. The same goes for the Baader-Meinhof German terrorists, and many other terrorist groups in different nations.

Examples such as these can usually be chosen to support a particular position on most issues, so more complete evidence is necessary to draw any conclusions with confidence. Fortunately, a few studies do systematically analyze the relation between poverty and terrorism. Harvard economist Alberto Abadie has recently studied both terrorism within a country and transnational terrorism for almost 200 nations (NBER Working Paper No. 10859). He estimates the poverty-terror relation after controlling for the degree of political freedom, religious and ethnic heterogeneity, and other variables. He finds little net relation between the degree of terrorism and poverty, where poverty is measured by per capita GDP, the degree of inequality within a country, and a couple of other ways.

Clearly, terrorism is important when there is political, ethnic, religious, and other conflicts between groups. Jewish terrorist organizations attacked the British army in Palestine, the Tamils oppose the Sinhalese and Moslems in Sri Lanka, the Moslems and Hindus of India and Pakistan continue to battle, the IRA has attacked both the British and protestants in Northern Ireland, and so on for many other examples. Abadie finds these connections as well as an important relation between terrorism and the degree of freedom. Countries with the greatest political freedom, such as Western Europe, the United States, and Japan generally have relatively little terrorism, although a number of exceptions include Germany's Baader-Meinhof terrorists and Italy's Red Brigades. Highly oppressive regimes effectively deter terrorism by close surveillance of their populations, and by severe punishments to apprehended members of terrorist groups. Countries in the middle ranges of political rights usually suffer the most from terrorism, perhaps because as Abadie conjectures, these countries are in political transition, with considerable disorganization and conflict.

Alan Krueger and Jitka Malechova examined the backgrounds of about 130 suicide bombers from Hezbollah in Lebanon who died on missions between 1982-1994. They found that these bombers were not poorer but rather were much more educated and better off economically than the general Lebanese population. One of Krueger's graduate students at Princeton found similar results for Palestinian bombers: a much smaller fraction came from poor backgrounds than is the fraction of poor in the Palestinian population as a whole, while more than half of all bombers went beyond a high school education compared to a small percentage of the Palestinian population who did.

I agree with the basic conclusions of the studies by Abadie, Krueger et al, and of a few others that poverty is not directly an important cause of terrorism. But a couple of significant qualifications are in order. The first is a technical point that may be crucial in interpreting some of the evidence. Any terrorist organization has available a supply of potential suicide bombers or other terrorists who have different levels of education and economic opportunities. To make my point in a simple way, suppose all potential bombers gain the same utility from destroying members of a hated group, such as Israelis, through successful attacks that are likely also to kill the bombers. Suppose too, they suffer to the same extent if they fail in their missions-they may be captured without hurting anyone, or they may only kill themselves.

Recruits with good economic opportunities would only be willing to undertake suicide missions that have a relatively high likelihood of destroying some enemies too. For they would not be willing to go on missions that have little chance of succeeding since they would then prefer safer terrorist activities, or doing well economically while working peacefully. In this case, relatively highly educated terrorists will be sent on missions that are more likely to succeed in destroying their enemies as well as themselves. As a result, the education and other determinants of the economic opportunities of successful bombers will exceed the opportunities of bombers who fail (and who may be captured).

Similarly, the education of captured bombers would be less than the education of all bombers since low educated individuals, such as the many teenagers sent on suicide efforts to Israel, would go on missions with smaller chances of succeeding. To protect against these misleading interpretations, the sample of bombers or other terrorists must be representative of all terrorists- not mainly either "failures" or "successes"- before reliable conclusions can be drawn about the relation between economic opportunities and the recruitment of terrorists. This analysis suggests that the Krueger- Malechova study of Hezbollah terrorists who died on their missions may be biased toward their finding that terrorism and variables like education appear to be positively related.

A second possible qualification would arise if the process of rapid economic development reduces terrorism by orienting more educated and abler individuals toward advancing economically rather than toward terrorist activities. I have not done a systematic study of the link between say economic growth and terrorism, but nations or regions that are experiencing rapid growth appear to have lower incidences of terrorism. Continuing economic growth also eventually leads to greater democracy, so a positive link between economic growth and democracy and a negative link between growth and terrorism could help explain the observed negative relation between terrorism and democracy.

To be sure, terrorism may be less common when nations are growing rapidly because the causation goes from terrorism to little growth; that is, terrorism discourages investments and other engines of growth. Whether the causation is from growth to little terrorism or from terrorism to little growth would have to be discovered from systematic and careful studies. But I believe that some of the causation runs from growth to reduced terrorism because it becomes harder to interest many individuals in risky terrorist activities (and other political activism) when economies are expanding rapidly and opportunities are booming.