Dividing the waters: An empirical analysis of interstate compact allocation of transboundary rivers

Legal scholars and jurists have identified several criteria (e.g., hydrology, climate, population, and historical water use) to guide equitable allocation of transboundary rivers among riparian claimants. Are these criteria used in practice, such that a quantitative pattern emerges from actual water-sharing agreements regarding factors affecting allocations? To address this, we study interstate compacts, the principal mechanism for allocating the waters of transboundary rivers within the United States. We develop a georeferenced data set and construct variables representing conditions in state-based watersheds of 14 rivers at the times of compact ratification. A state's water allocation share of a compact serves as the dependent variable, and a set of explanatory variables is derived from legal and political theories. We estimate allocation shares using both ordinary least squares (OLS) and bootstrap regressions, and we apply two alternative specifications of the factors affecting compact allocations, one with and one without political variables. Estimated coefficients on variables for land area, population, prior water use, riparian position, and Congressional committee chair are statistically significant in the OLS regressions. The preferred OLS specification, which includes political variables, provides a good fit (R ² = 0.84). We also find that OLS and bootstrap regressions have a similar ability to predict state allocation shares. We discuss how the results could be used as a reference point in negotiations over new compacts or international river treaties and as a basis to identify existing compacts with statistical outliers.