A quantitative analysis of private property rights in groundwater

Economists have long maintained that when a groundwater resource is common property, stock externalities induce an inefficient rate of groundwater pumping. The remedy usually prescribed is central control by a 'water czar', who uses taxes or quotas to obtain the efficient allocation of the resource over time. This thesis examines the welfare consequences of an alternative economic arrangement, in which private property rights to the groundwater stock are established and the market for these rights controls the aggregate rate of groundwater pumping. The thesis is divided into two chapters. The first develops an analytical framework for comparing the rate of groundwater pumping under the private property rights regime to the rates arising under the common property and central control regimes. The rate of groundwater pumping under the common property regime is cast as the equilibrium solution to a noncooperative dynamic game. The rate arising under the private property rights regime reflects a rational expectations equilibrium. Analytical results demonstrate that when firms are risk neutral, the private property rights regime provides a welfare improvement over the common property arrangement, but in the general case it falls short of providing the welfare gain attainable by a well-informed water czar. On the other hand, when firms are risk averse, the private property rights regime may provide greater welfare than central control.;The extent to which the private property rights regime fails to achieve the efficient allocation of groundwater when firms are risk neutral is the empirical question addressed in the second chapter. A stochastic dynamic programming model of Madera County, California, is developed. To avoid the 'curse of dimensionality', an approximation method is used to derive the optimal pumping policy. Programming results suggest that privatizing the groundwater resource in Madera County would yield 70 to 85 percent of the welfare improvement possible under central control. Significantly, results indicate that the welfare improvement from central control would be small.