Market power in California's water market

Article  -  Monday, 3 June 2024

Tomori, F., Ansink, E., Houba, H., Hagerty, N. and Bos, C. (2023): "Market power in California's water market", American Journal of Agricultural Economics

The authors study the extent and impact of market power in water markets. Such markets are not abundant globally but their prevalence has been increasing. They use a Nash-Cournot model and derive a closed-form solution for the extent of market power in a water market setting. The authors then use this solution to estimate market power in a newly assembled dataset on California's statewide surface water market. California is one of the world's largest water markets by quantity and value of water traded.

The model starts with two main assumptions. One assumption is that they fix the side of the market on which market power resides. Their starting point is buyer-side market power. Authors then estimate the model in the other direction, allowing for seller-side market power, and find none. To check the relevance of this assumption, they also employ a model specification where they allow for market power on both sides at once; they find compelling empirical support for market power on the buyer side only. The second main assumption is that authors use linear demand, originating from a quadratic benefit function of water use. This functional form is commonplace in the water economics literature and allows for a straightforward empirical strategy to derive our results. Constant linear demand across selling districts may not be realistic, however, and therefore they relax this assumption in an alternative specification where, instead, they impose a constant price elasticity. This alternative specification, with constant price elasticity, is presented as part of a larger class of model specifications featuring homogeneous demand, for which they present a closed-form solution as well.

Their main result is that market power in California's water market is limited. The main specification implies that buyer power yields an average markdown of 6% of the transaction price. This result is obtained for the linear model but continues to hold for the nonlinear specification and is robust to other model modifications. It is surprising in the sense that the thinness of water markets, including California's, is conventionally associated with higher possibilities of exploiting market power. Intuition for this result may be found in the idea that an individual buyer rarely purchases a large fraction of any seller's endowment of water. Each seller not only has many potential buyers but also consumes water directly, leaving residual supply highly elastic to each buyer.

This study suggests that market power is not a first-order concern for policy reforms aimed at improving the efficiency and flexibility of water allocation in California, and perhaps in other water-scarce contexts as well. Reform efforts can focus on other considerations instead, such as transaction costs and political economy factors. Proposals to, for example, break the control of water districts over water rights may or may not be desirable on other grounds, but market power in the statewide water market does not appear to be one of them.

Up