How do demographic factors such as diversity influence the success or failure of administrative decentralization programs? This project investigates the impact that the establishment of Water User Associations (WUAs) had on the equity of irrigation water distribution in the Jordan Valley. Rather than arguing that the decentralization program itself succeeded or failed in its attempt to improve water management systems, I argue that this effect is moderated by the demographic characteristics of each area. Based on previous literature that has described a variety of opinions concerning how helpful these types of associations have been, I show that such decentralization programs are most successful in areas where the potential for corruption based on kinship ties is lowest. In areas where that potential is higher, we can expect to see unintended negative consequences on water management systems in that the institutions became vehicles through which existing elites could legitimatize their authority with little oversight. Throughout this project, I make the argument that we are most likely to see corruption when there is large variation in the relative ability of individuals to engage in corruption based on familial and tribal ties. In these cases, individuals are able to participate in corruption networks at the expense of their neighbors. In other cases, where the relative ability to engage in such transactions varies little from person to person, the total level of corruption falls, and we would expect a more equitable distribution of water. In the latter case, we would expect to see a positive effect of the introduction of local governance institutions. In the former, we can expect to see either no effect of these institutions or a negative effect wherein the establishment of WUAs facilitates corruption. I provide evidence for my argument through a combination of original interviews with farmers and personnel from the Jordan Valley Authority conducted from January 2024 through October 2024, quantitative analysis of differences in crop production within the Jordan Valley, and a novel dataset that uses voter registry data and tribal genealogies to develop estimates of tribal diversity at a local level. Tying together the political science and economics literature on decentralization, previous work on distributive politics and determinants of public goods provision, and anthropological accounts of Jordan’s tribal dynamics, this project contributes to our understanding of how existing social structures and hierarchies influence the effect of decentralization programs, challenging common one-size-fits-all approaches to development.