The Policy Implications of Using Water More Than Once - The Case of Irrigation in Egypt

When water is scarce and agricultural production is dependent on irrigation, the question naturally arises how to allocate water most efficiently from society’s point of view. Standard volumetric water pricing targets the amount of water the farmer applies to his field. However, what matters from a social efficiency point of view is often not how much water is applied to the field but rather how much water is consumed by the crops and evaporation, as a substantial amount of the water losses due to drainage and percolation tends to be recycled in a basin and become available for additional application cycles. One example of this ability to use water more than once can be found in the Egyptian agricultural sector. Egyptian agriculture is traditionally centred around the Nile, and in the Nile Valley virtually all water losses are recovered by the basin water system. So far Egyptian farmers have not had to pay for their irrigation water, but water scarcity is expected to be a growing problem in Egypt due to population growth and reclamation of desert land for agricultural production. Efficient allocation of irrigation water is consequently becoming an important issue from an economic efficiency point of view, and the fact that water losses are at least partially recoverable has implications for the socially optimal choice of policy instrument to regulate the farmers’ use of irrigation water. Based on results from a theoretical model I have recently developed on the optimality of policy instruments for regulating farmers’ use of irrigation water, the present paper will explore the empirical implications of different direct and indirect water allocation policy instruments ranging from water pricing to land taxation. The theoretical analysis will be incorporated into a modified version of the ORANI-G CGE model and the IFPRI 1997 SAM for Egypt, where the use of water in agriculture has been introduced. The aim of the analysis is to compare the production and welfare implications of the theoretical first best instrument of combined volumetric water taxes and return flow credits with the theoretically second best instruments of standard volumetric water taxes or crop-specific land taxes. Simulation results for production and welfare implications are reported for the two kinds of water tax policy instruments, but a satisfactory implementation of crop-specific land taxes have not yet been found. It is nonetheless expected that crop-specific land taxes will perform relatively well from an efficiency point of view.