Do different types of military alliance commitments have different impacts on interstate disputes? In this dissertation, I test the effects of different alliance commitments on interstate disputes, focusing on comparing the effects of having a defense pact and those of having a consultation pact. Another contribution of this dissertation is to address two methodological issues relevant to the research question, endogenous treatment and sample selection bias. Knowing that states choose alliance commitments in the prospect of interstate disputes, one should consider the alliance choices as a treatment variable that cannot be randomly assigned. Furthermore, in studying the latter phases of interstate disputes, the sample of the subsequent decisions is not representative of the entire population unless interstate disputes randomly occur, which is hardly justifiable in the International Relations literature.

The three main chapters of this dissertation can be best described as a demonstration of my learning process. Each chapter shares the same agenda described above but tackles different sets of the problems. Chapter 2 examines the effects of a consultation pact and those of a defense pact on the other states' decision to attack an alliance member, the alliance member's decision to attack others, and the alliance member's decision to respond militarily when attacked. However, this chapter does not address any of the two methodological challenges described above. Chapter 3 examines the effect of a defense pact on the first and the third decision using a switching probit model, addressing endogenous treatment issue. Chapter 4 addresses all of the research agenda mentioned above, except examining the second foreign policy decision. I develop a Bayesian multiple equation model specifically tailored to testing the effects of military alliances and use the model to study the effects of alliance choices on interstate disputes. In conclusion, by adopting several different approaches to the same research question, I find that a defense pact is negatively associated with the probability of being a target of interstate disputes. On the other hand, it is not very clear how other alliance commitments are related to interstate dispute initiation and escalation.