IWRA Proceedings

< Return to abstract list

Who gets what? policy lessons on how to share scarce irrigation water during droughts

IWRA World Water Congress 2008 Montpellier France
6. Water Conservation and Demand Management
Author(s): Sophie Thoyer
Deborah Peterson
Jean-Michel Salles


Keyword(s): Water availability, use and management
Article: PDF

AbstractWith climate change, a number of regions in the world are forecast to face more frequent and more severe droughts, forcing the irrigated agricultural sector – often a major consumer in summer periods- to face water restrictions. One of the main challenges for policy-makers is to design mechanisms for sharing water among farmers, which can be rapidly and easily enforced, while ensuring that water is efficiently allocated without creating unnecessary inequity. Where there are no markets for water, a regulatory approach to managing demand for irrigation water is often employed – for example, irrigators are permitted to water only on certain days. However, between competitive markets and regulatory individual restrictions, there is a wide range of other solutions. This paper examines six alternative approaches to managing irrigation water in drought periods when water markets are not available or when farmer entitlements are not defined as a percentage of shared resource: shortage pricing; public buy-backs of water entitlements public purchase of reductions in irrigated land; water options; interruptible contracts; and instruments of collective reward or punishment. The efficiency, effectiveness and feasibility of each instrument is considered and compared with the regulatory approach. Case studies, where examples exists, in Australia, USA, Spain and France are analysed. The paper focuses exclusively on the policy solutions dealing with conjunctural, unforeseeable deficits, associated with an exceptionally dry season and low reserves. It therefore has to take into account the short-term constraints of farmers who may have to face restrictions at short notice. A simple microeconomic farm model is built – with stochastic water supply and a risk-averse farmer. Based on this model and under different scenarios of water supply (with or without access to groundwater, and varying time of restrictions), the impact of the six instruments on farmer’s irrigation decisions, expected farm revenue, and water use is examined as well as the impact on budget costs and total agricultural water use. A comprehensive literature review of theoretic models and applications facilitates the comparative analysis. While a regulatory approach, depending on its design, can offer advantages in terms of low administrative transactions costs, it offers the least flexibility to farmers and hence reduces economic efficiency. Pricing approaches, such as a uniform volumetric price or block pricing, requires ability to measure volume or water used. Increased charges for water will have no impact on quantity of water used if demand in the relevant price range is already constrained; in other words, the price needs to be sufficiently large which may limit the political feasibility of this approach. Buy-backs of water entitlements and paying farmers not to irrigate reveal information about the opportunity costs of water to irrigators and allow efficiency gains relative to regulatory approaches. However, design of such programs is important and direct costs to governments can be high. An options contract written against irrigator’s seasonal allocations of water could reduce water used at a significantly lower cost than purchasing entitlements. By retaining their entitlements, irrigators have a hedge against the risk of decreased water availability in the future. Interruptible contracts, a concept used in electricity markets, allow revelation of the value of supply interruptions to individual farmers. Specification of the contract details is a significant challenge. If farmers in the affected region are relatively homogeneous, the costs to water managers may be high. Instruments of collective reward or punishment can be useful when individual uses are not easily measured but they are not easily accepted.