Policies and technologies for a sustainable use of water in Mexico a scenario analysis

This dissertation studies the role of public policies in promoting the adoption of alternative irrigation technologies to support a sustainable pattern of water withdrawals in Mexico. The study implements a multi-factor, multi-sector, multi-region model of the Mexican economy for thirteen hydro-economic regions to capture the main features of the interface between water resources and economic activity. The model allows for the choice among alternative agricultural technologies in each region and solves for cost-minimizing allocations of output that are constrained by regional factor availability. The model also features an endogenous mechanism that determines the magnitude of commodities prices based on production costs and of scarcity rents of limiting factors of production.;The study designs five scenarios combining the implementation of water policies and the availability of alternative irrigation technologies to test three hypotheses. A baseline scenario computes water withdrawals in the absence of water policies while allowing for the choice among the two agricultural technologies that are dominant in Mexico: non-irrigated and food-irrigated agriculture. A sustainable scenario simulates the enforcement of region-specific quantity restrictions on water withdrawals to meet sustainability criteria. The scenario tests a first hypothesis regarding the physical feasibility of water sustainability with baseline technologies, which the study fails to reject. The sustainable solution requires a decrease of 10% in the national use of irrigated agricultural land and an increase of 10% in the use of non-irrigated agricultural land. Since the former is found to be more cost-effective than the latter, the sustainable solution is characterized by an increase in the price of agricultural output of 36% relative to the baseline. In the sustainable scenario, sustainable endowments of water constrain economic activity in five regions, comprising the country's North and the Mexico City Valley, for which positive scarcity rents for water arise.;Although the study recognizes that the sustainable solution can not be improved upon via the imposition of exogenous prices for irrigation water, for such a solution is cost-minimizing and water-sustainable, a third scenario studies the impact of imposing such prices in unsustainable regions with baseline technologies. The scenario tests the particular policy of imposing water prices equal to the water's shadow prices found in the sustainable scenario, which is found to fail in promoting a sustainable pattern of water withdrawals. The fourth and fifth scenarios expand the set of agricultural technologies by including a mix of sprinkler and drip irrigation systems, and test the imposition of quantity restrictions and water prices, respectively.;The costs of technology adoption in these scenarios are separated between operating costs, which are implemented endogenously into the model, and upfront investment costs, which are assumed to be subsidized and, therefore, held exogenous to the model. These scenarios fail to reject a second hypothesis stating that public policies can promote technology adoption, support sustainability of water withdrawals, and generate agricultural prices lower than in the sustainable scenario. Pricing policies, on the one hand, are found to promote adoption in 31% of used irrigated land and to produce an increase on agricultural prices of 5% relative to the baseline. Quantity restrictions, on the other, are found to promote adoption in 24% of used irrigated land and to produce an increase in agricultural prices of 8% relative to the baseline.;The third and last hypothesis states that these public policies can potentially generate enough water revenue to cover for the total costs of subsidizing technology adoption. A wide value range for upfront investment costs is constructed from a database available in the literature that contains estimates from hundreds of cases around the world. Depending on the degree of technology adoption for each technology scenario, corresponding annuities are calculated showing their sensitivity to the choice of different time-horizons (five, ten, and twenty years) and discount rate (5% and 10%). The exercise finds that only when the upfront costs lie at the higher end of the estimate range, and when 5-year time-horizons are considered, is the goverment's water revenue not enough to fully cover for total adoption costs. At the extent to which these estimates are at the middle or at the lower end of the value range, or time-horizons of 10 years or longer are considered, can the government generate enough water revenue to fully cover for the total adoption costs. Therefore, the study fails to reject the third hypothesis.