Generalised social norms in water consumption networks

According to the last IPCC report (AR6, 2022), 'roughly half of the world's population are assessed as currently subject to severe water scarcity for at least some part of the year due to climatic and non-climatic factors, and this is projected to be exacerbated at higher levels of warming'. Due to this rise of the water scarcity, there has been an increase of small-scale systems managed by local authorities or companies such as smart meters which provide information on similar household consumption to help consumers reduce their water use. Many theoretical papers use the average consumption of others to define the social norm information provided to agents (Ushchev and Zenou 2020; Pedehour and Richefort 2022). New to this approach, we use a more general definition of the social norm following the recent paper of Boucher et al. (2024) who consider that social norm can be based on other measures. For example, median or minimum consumption could be used as a new reference point for water consumers. In this context, this paper develops a model of water extraction with endogenous social norms to study the different effects of this informaƟon provision and calibraƟon on the networks of water consumers. We consider a territory where water consumers are connected to a unique shared resource that can be overexploited in case of water scarcity. Preferences of individuals are guided by their extraction values and their taste for conformity to follow their norm. Depending on the calibration of the norm, the position in the network of agents, and the connections between water consumers, some definitions of social norms appear more optimal than others for reducing water consumption. We find conditions for uniqueness and interior solutions of the equilibrium. We then establish welfare and social-opƟmal properties to characterise social norms that could avoid the tragedy of the commons and suboptimal water consumption. Preliminary results highlight the strategic interaction between consumptions and can be used to derive interesting policy implications about the optimal norm information to set.