New Models of Multilateralism: Investigating the Dynamics of Transboundary Water Governance

Although there is considerable research analyzing the determinants of river basin treaty formation, accurate modeling of multilateral river basins remains a challenge and their dynamics are incompletely understood. Despite being theoretically and empirically expedient, dyadic data may not be the best method for investigating multilateral systems. This research aims to fill the gaps in multilateral river basin analysis and provide a new approach to model constrained multilateral systems.

This dissertation investigates three aspects of multilateral transboundary water governance and is organized in a journal paper format. The first paper revisits a published study that made a significant contribution to the transboundary water literature, but makes a substantial change to the data structure to test the robustness of the original results to a change in unit of analysis. Results show that dyadic data analysis of multilateral interactions is sensitive to a change in unit of analysis that accounts for all members of the multilateral system.

The second paper develops an analytical solution to the data structure problem by combinatorically expanding the dyadic data to create an all-inclusive counterfactual of all possible treaty-making combinations of member countries in each river basin. This unique dataset accounts for all treaty possibilities: bilateral, plural-lateral, and multilateral. Furthermore, this new dataset offers a reasonable solution to the common pitfalls of dyadic analysis of multilateral systems. Analysis shows that the counterfactual data offers the most predictively accurate modeling of multilateral treaty making in river basins.

The third paper addresses the most substantial difference in results from the unit of analysis change considered in the first two papers. Contrary to popular international relations theory, power does not seem to have a significant effect on river basin treaty formation. Believing this may be due to measurement error, this paper constructs a new measure of power specific to the dynamics of multilateral river basins. It then tests the new measure for efficacy.

These three papers as a whole address the problem of modeling multilateral events in river basin research and show that careful modeling of multilateral systems offers greater understanding of the dynamics of riparian decision-making. More generally, these studies offer a blueprint for solving a common problem in international relations research that involves constrained multilateral systems, i.e., multilateral systems that have defined participants and cannot randomly add participants.