In this paper we consider the uses political scientists have made of agent-based modeling (ABM) and the challenges associated with designing research at differing levels of complexity. We propose a typology of ABM research designs-investigating abstractions, testing theories comprised of ensembles of simple variables, or implementing virtualizations of complex situations. Our illustrations are drawn from work done on problems pertaining to the evolution of collective identities and norms and to their contribution to collective action. By increasing the visibility of research design questions and clarifying the choices and opportunities associated with them, we seek to expand the scope of responsible methodological uses of ABM techniques and render the increasing variety of that work accessible to wider audiences.
This article examines how the relationship between ethnic polarization and civil war could be moderated by different degrees of ethnic salience. Using an agent-based computational model, we analyze the polarization-conflict relationship when ethnic salience is ``fixed''-high for every member of two nominally rival ethnic groups and ``variable''-permitted to vary across individuals as a function of relative income. We find that (1) when salience is fixed, conflict onset is twice as high at low levels of polarization compared to when salience is permitted to vary, with the difference decreasing at high levels of polarization; (2) the relationship between conflict onset and the range over which we calculate variable salience is positive and robust for low and moderate levels of polarization; (3) the relationship between polarization and conflict onset is robust even under minority domination, if one holds salience fixed; and (4) holding ethnic salience fixed effectively amplifies the negative effect of polarization on economic performance.
The emergence of an ever-widening sphere of global public policy is a new reality in a world characterized by the blurring of boundaries between the national and the global; by flows of ideas, people, and commodities; and by new global risks and opportunities. In this context, this article explores the empirical puzzle of the sudden outbreak of reforms leading to central-bank independence. How can we best understand the outbreak of reforms in the 1990s? It is suggested here that the reforms were diffused in a contagious and uncoordinated manner in a global policy process that may best be captured by Kingdon's policy stream model. We develop an agent-based model to evaluate the effects of three little-explored aspects of the diffusion process. These are (i) the likelihood of the outbreak of reform, (ii) the rate of adoption of the reform, and (iii) the time to outbreak. We find that the likelihood of outbreak depends on the saliency of a problem, in conjunction with the length of time that a problem has been on the public agenda. We also find that an increase in the size of the environment surveyed before a decision is made increases the rate of adoption but also the time to outbreak. The more global the information available for agents, the longer is the time to outbreak, but outbreaks unfold much faster.