The problems of water economics, I submit, are more those of the organization and management of self-supplying firms and of governments at all levels than those of an industry as the term is used in economic theory.I found this (i.e., the dominant paradigm is bureaucracy, not markets) to be true when writing my dissertation.
After a wide-ranging discussion of elasticity, demand projections and social institutions, Wantrup goes on to address the "misallocation" of water to agriculture. Except that he doesn't call it misallocation, i.e.,
It is true that the statutory preference given to agricultural over industrial use is obsolete. I have gone on record to that effect on previous occasions. I suggested then to eliminate statutory preference altogether and to leave to the courts or special water rights boards the determination of which is the higher use in each situation of conflict. On the other hand, the economic significance of agricultural preference and priority is already limited by other provisions in western water law. There are no less than seven of these that are relevant.So perhaps agricultural allocations reflect the balance of social objectives.
While that may have been true in 1961, it's hard to say that it's true today. Even farmers complain that the allocation of water rights within agriculture does not even get close to efficient.
He expands on this theme (optimal water allocation) in a 1967 paper [pdf] in which he points out that water laws and institutions [at the second level] should provide a flexible framework for [first level] allocation decisions that might change with circumstances, i.e.,
The purpose of decision making on the second level is not to control directly inputs, outputs, and other quantitative characteristics of the water-resources system nor to obtain a path of quantitative welfare optima at various points in time under projected conditions for these points. Rather, the purpose is to maintain and to increase welfare by continuously influencing decision making on the lower level under constantly changing conditions that for any point in time cannot be projected -- or only vaguely -- and that are always uncertain with respect to actual occurrenceThe paper goes on to criticize economists who make unrealistic assumptions that enable them to calculate short-term optimal allocation while ignoring the "survival value" of the system over time.
Bottom Line: Listen to Wantrup! Things are more complicated than they appear.