RURDS Vol. 19, No. 2, July 2007 doi: 10.1111/j.1467-940X.2007.00128.x MODELING THE ECONOMIC EFFICIENCY OF IRRIGATION WATER SUPPLY IN DRY AND HUMID REGIONS Yeti Nisha Madhoo Department of Economics, University of Mauritius, Le R´ eduit, Mauritius This paper tests the economic efficiency of irrigation water as supplied by single purpose and by general purpose public utilities in dry and humid regions, respectively, in terms of underprovision and overprovision. The proposed methodology mixes conceptual frame- works developed by Farrell (1957) and Brueckner (1979, 1982). A typical agricultural production function uses the institutional attributes of the water-providing authority as an argument. Both short-run and long-run estimates are provided. Although in both cases irrigation water is heavily subsidized, there is no evidence of systematic underprovision or overprovision for the dry north and west, under the single purpose Irrigation Authority. However, water is oversupplied to more humid areas by the general purpose Central Water Authority. Empirical estimates confirm that the efficiency of the single purpose Irrigation Authority cannot be attributed to irrigation technique. The general purpose Water Authority suffers from inefficiency in coordinating rainwater availability with the institutional water supply. 1. Introduction There is no dearth of literature on inefficiency in public sector enterprises and misman- agement of government budgetary deficits in developing countries (for instance, Compos and Esfahani, 1996). However, empirical measurement of inefficiency in irrigation water supply by public utilities has not been attempted systematically. The crux of the present research is to determine whether publicly provided irrigation water adds value to economic activity where it is used. An adverse impact on production may be felt both due to underprovision and overprovision of water. Therefore, the economic inefficiency of water supply can be measured in terms of either its underprovision or its excessive provision. Since water is subsidized, demand for this good is artificially high and public utilities generally over-provide to some groups. Overprovision of irrigation water would be followed by another kind of inefficiency: water would become less available to other user categories, which would then be underprovided. In both cases, output effects would be negative. Besides misallocation of irrigation water across uses or users due to subsidies, a lower price of water as an input may generate profit effects having a positive impact on output (see Singh et al., 1986; Taylor and Adelman, 2003 for summary). This would mean that subsidized water can have two effects on output: a negative, inefficient provision effect and a positive input price effect. Inefficiency can further be attributed to climatic conditions, technological differences, and coordination failures. Irrigation technologies have taken such forms as overhead and drip C  The Applied Regional Science Conference (ARSC) / Blackwell Publishing Asia Pty Ltd. 2007. Published by Blackwell Publishing Asia Pty Ltd.