The case study conducted in this paper looks at residential water pricing from three different points of view. It first describes existing urban water-pricing practices in Southern France, emphasizing that pricing is not yet being used as a tool for providing economic incentives to save water. It then looks at the observed impact of pricing on water consumption, through an econometric analysis of a cross-sectional data set. The analysis suggests that demand, with an estimated price elasticity of -0.2, is not yet very responsive to price variation. A regional water model (300 municipalities) is then developed and used to simulate the potential impact of various water-pricing scenarios on aggregate water demand, aggregate water sales revenue, and consumer surpluses. The results illustrate the trade-offs that have to be made between the search for environmental effectiveness, cost recovery, and equity when implementing complex water-pricing structures such as block rates or seasonal water pricing.

The present work focuses on the demand side of future water scarcity assessment, and more precisely on domestic water demand. It proposes a quantitative projection of domestic water demand, combined with an original estimation of the economic benefit of water at large scale. The general method consists of building economic demand functions taking into account the impact of the level of equipment, proxied by economic development. The cost and the price of water are assumed to grow with economic development. The methodology was applied to the Mediterranean region, at the 2060 horizon. Our results show the evolution of water demand and value, measured by surplus, over time. As long as GDP per capita and water price remain low, demand per capita increases along with economic development, and surplus per capita increases with demand. As demand approaches saturation, the combined negative effects of water cost and price increase on surplus grow stronger, and surplus per capita begins to decrease. The developed methodology is meant to be used for large-scale hydroeconomic modelling, in particular for regions with heterogeneous levels of development and low data-availability. (Résumé d'auteur