The Rational Design of Security Institutions: Effects of Institutional Design on Institutional Performance

Abstract

Based on the assumption that security institutions are designed rationally, I study the variations in design schemes and their possible effects on institutional performance. Military alliances vary with respect to their membership size, level of security obligations undertaken by the allies, incorporation of non security clauses such as economic agreements, level of institutionalization, specified duration of existence, as well as the conditions under and reasons for which they are formed. This dissertation studies the effects of above mentioned design features on the probability of security alliances expanding their scope by addressing non-security agreements such as free trade agreements and conflict management clauses. I find support for the argument that states include economic agreements within a military alliance as a means to bolster the credibility of an otherwise weak security alliance. Results indicate that allies facing high levels of external threat and low levels of intra alliance cohesion are more likely to include conflict management provisions in the alliance. Finally, I conduct a systematic study of the possible effects of variation in structural design on the durability of an alliance. I find that design features that increase the costs of breaking the alliance increase the duration of an alliance.