Here is the abstract:
The Metropolitan Water District of Southern California (MET), a cooperative of retail and wholesale water utilities, serves 18 million people. This case study explains how MET---as a cooperative---is inefficient and how its member agencies suffer from this inefficiency. I show that MET is inefficient by demonstrating that its members have heterogeneous preferences over outcomes: Members that are more dependent on MET prefer policies that increase water supply; others prefer lower rates. Although heterogeneity had existed since at least the 1940s, MET avoided conflict well into the 1970s. I explore two possibilities for efficiency despite heterogeneity. First, MET had so much water that it could treat it as a club good, i.e., members did not need to agree on policies over non-rival water. Second, member agencies may have had social preferences ("one for all and all for one"). Shrinking subsidies and supplies in the 1960s changed water from a club to private good. The end of social preferences is not so obvious, so I asked MET's member agency managers to participate in public goods experiments. They do not appear to have social preferences. If MET is inefficient as a cooperative, we should see evidence of this inefficiency, and MET's pricing policies (setting annual prices in the prior year and selling water for the same price to all locations) provide this evidence. With increasing water scarcity, the damage from these policies is growing.Bottom Line: This dissertation was a lot of work, but it was worth it.
I use 60 years of panel data to show that water increases land value, dependency lowers it, and water may have been misallocated during the 1987--1991 drought. I describe how marginal water can be auctioned after inframarginal, "lifeline" water is allocated and present experimental results for "water" auctions in which water managers suffer endowment effects but compete more (relative to students).
In addition to the analysis of MET, other contributions are a quantification of bargaining power within an organization (dependency), measurement of water manager cooperation, estimation of the value of water on urban land, and design of auctions for equity and efficiency.