education, poverty and lack of economic opportunity are positively correlated with
terrorism. In the case of our model, a formal presentation is required because the
interactions are simply too complicated to decisively understand in an informal
analysis.
Previous formalized rational-choice analysis of terrorism has focused on two
areas: (1) the interaction between governments and terrorist groups; and (2) groups’
internal dynamics. In the first area, much of the early work focused on the signaling
dynamics of terrorism. Lapan & Sandler (1993) present a model in which the optimal
government strategy depends on the resources of the terrorist group. The terrorist
group can use the scale of attacks to send a signal about these resources and has
incentives to misrepresent its resources to gain concessions from the government.
Overgaard (1994) presents a more subtle signaling model in which terrorist resources
are renewable between periods and the terrorists have positive alternative uses for
resources, alternatives such as providing social services. Overgaard finds that if
concessions are ruled out for exogenous political reasons, then only a pooling
equilibrium exists and the government learns nothing from the scale of attacks. More
recent work examines the operational competition between terrorists and the
government. Sandler (2003) explains the under-provision of international counter-
terrorism. Frey & Luechinger (2004) use a simple supply and demand model to
examine the relative merits of deterrence and decentralization of critical infrastructure
as counter-terror policies.
In one of the first formal rational-choice analysis of terrorist groups’ internal
dynamics, Chai (1993) examines why people participate in covert anti-government
organizations even though their chances of getting particularistic benefits out of doing
so are quite low given the low likelihood of overthrowing the government. His
4
Convention