Polycentricity through policy layers: what connects perceptions, financing, and laws of water-uses relationships?

governance systems are often complex due to the number of interactions between these centers and actors, wildly when they are conflicting and overlap. In the case of water, this number of interactions is expanded because water uses are multiple and interdependent. Consequently, water governance became a privileged case for studying polycentricity and handling its complexities (Carlisle and Gruby 2019; Lubell, Robins, and Wang 2014; Ostrom 1990). Most studies focus on multi-sectoral, multi-level, or spatial aspects. But governance is also polycentric because combining different policy process layers, the most general being politics, policy, and polity (Hill and Varone 2014). This paper aims to characterize how these layers articulate in a complex polycentric setting and the factors affecting this articulation. We consider multiple water uses, define complexity as a function of the number of interactions, and focus on three intricated governance layers. We assume that coordination effectiveness depends on the synchronization of the interactions among water uses within and across the governance layers (Teisman and Edelenbos 2011; Renou and Bolognesi 2019). We focus on three layers that are representative of the policy process. The first layer is the perception of interconnections intensity between water uses because perceptions are an important determinant of policy agenda by affecting political demand, policy acceptability, and policy-problem definition (Jones and Baumgartner 2004). The second layer is the financing flows, as they are a critical policy resource (Knoepfel et al. 2011; Lambelet 2019). Through formal laws, the third layer is institutions, as they shape and contribute to defining the actors' behavior. The water management in the Geneva Canton serves our empirical inquiry. We define 19 different uses, i.e., 340 possible relations among water uses. We surveyed stakeholders to assess the perception of each relation. Then, we measure the financial flows linking each use and report each flow's legal. We use network analyses to delineate the complexity of the system. Results indicate what favors coordination and where are the most significant lack in coordination. They contribute to advance knowledge about what shape polycentric design and policy integration in complex coordination settings.