Levi Crews - Princeton University

The agricultural sector is the dominant user of water, but its use is distorted by two key forces: (i) farmers’ imperfect property rights over water extraction and (ii) the various taxes and subsidies on their products. Using a rich collection of detailed geospatial data, we document that these patterns are pervasive across the globe. We then build a dynamic spatial general equilibrium model to quantify the potential welfare gains of reallocating global water use in agricultural production. In the model, each local water stock evolves endogenously as nearby farmers extract water as if from a common pool. Farmers’ choices of which crop to grow and how much to produce respond, in turn, to the level of the local stock and the prices they face in the global market. We quantify the model such that it rationalizes observed patterns in agricultural production and trends in water availability out of steady state. With the quantified model in hand, we consider how counterfactual agricultural policies shift water use and affect welfare in the long run.