This award was funded through the Social and Behavioral Dimensions of National Security, Conflict, and Cooperation competition, a joint venture between NSF and the Department of Defense.
Policy makers need to understand dictatorships better in order to craft effective foreign policy, but systematic research that could result in more reliable predictions about dictatorial behavior in different situations has been limited by inadequate data. Scholars who study autocracies agree that they differ from each other in terms of who has the power to make decisions, which elites and societal groups have influence on these decisions, whether institutions constrain arbitrary dictatorial behavior, how easy it is to remove leaders, and the cost of ouster to leaders. They also agree that these differences should be expected to affect international and domestic behavior. In order to explain and make more reliable predictions about dictators' behavior, analysts need information that allows them to make distinctions among different kinds of autocracy because dictatorships led by different actors and responsive to different elite and societal groups behave in different ways. As part of this project, the PIs will gather information about approximately 175 authoritarian regimes. For each regime, about 60 characteristics relevant for explaining or predicting authoritarian behavior will be coded (or updated and checked for accuracy) and made publicly available.
Analysts also need to be able to estimate dictators' time horizons if they are to predict their behavior. Time horizons influence key policy choices ranging from the provision of public goods at home to the initiation of conflict abroad. Dictators who expect to be in power for a long time have reason to invest in policies that contribute to economic growth, in contrast to those who feel threatened by imminent ouster. Threatened dictators also tend to engage in rash international behavior. The project will generate measures of authoritarian time horizons. Time horizons will be modeled by estimating the predicted probabilities of individual dictator ouster and authoritarian regime failure. Preliminary analysis shows that both depend on specific characteristics of the dictatorship in question, e.g., a "sultanistic" dictator like Saddam Hussein has a different time horizon than does the ruler of a party-based regime like the Chinese.
