California experienced a long period of drought between 1987 and 1992. This situation became severe in 1991, and action became necessary in order to allow for more efficient water distributions. Thus, in 1991 and then again in 1992, the California state government set up Drought Emergency Water Banks to facilitate the allocation of water. These Banks were significant in that they were the first large water transfer programs in the US set up and run by a state government. The basic format was essentially the same for both Banks, though they performed quite differently.

In both 1991 and 1992, the Department of Water Resources(DWR) was given the task of establishing and operating Drought Water Banks. They served as a broker for water transfers. Specifically, they drafted contracts for the purchase and sales of water, and performed as the agent through which transfers passed.

In the case of the 1991 Bank, the DWR set a fixed price for both purchases of water by the Bank and sales of water by the Bank. The purchase price of water was set at $100/1000m3, and the Bank bought 1,013,000m3. Half that water came from farmers who fallowed their land and sold the water their crops would have consumed to the bank. One-third of the water came from farmers who substituted surface water rights with groundwater rights and sold the surface water to the Bank. The remaining water came from surface water supplies in Northern California that had excess water.

The sale price of water was set at $140/1000m3 by the DWR, and the Bank sold 488,000m3. About 80% of the water sold by the 1991 Bank went to urban municipal and industrial uses. The remaining 20% was purchased for agricultural use. The difference between purchase price and sale price represented the cost of water carriage. The difference between quantity purchased and quantity sold can be accounted for in the following way: In order to maintain water quality standards, the Bank had to release 20-30% more water into the system than it removed; the remaining surplus, 37% of the total water purchased, was a result of poor planning, and was purchased from the bank by the DWR as a backstop.

As a result of lessons learned from the 1991 Bank, the 1992 Drought Emergency Water Bank functioned more efficiently than its predecessor. The surplus of water that resulted in the 1991 Bank led the DWR to change the contract system. Specifically, a dual-contract system was implemented  Under this system, buyers and sellers who committed to the Bank early made a deposit on the water they thought would need or would be able to supply. This allowed the Bank to anticipate supply and demand more accurately. Thus it was able to avoid much of the excess supply that existed in the 1991 Bank. Other measures intended to limit the surplus in the 1992 Bank included much lower purchase and sale prices and a ban on water from fallowed land.

The 1992 Drought Bank purchased water at a price of $40/1000m3. They bought only 232,990m3 of water, 23% of that purchased in 1991. 80% of this water came from farmers who substituted surface water with ground water. The remaining 20% came from surpluses in Northern California surface water supplies.

The Bank sold all the water at a price of $58/1000m3, though this price was not officially fixed. Unlike the 1991 Bank, the majority (60%) of this water went to agricultural uses. The Department of Fish and Game bought 15% of the water to account for undervalued instream uses, and the remaining 25% went to urban uses.

Many conclusions can be drawn about water transfers and water allocations in light of the California Drought Emergency Water Banks of 1991 and 1992. The Banks were essentially institutions designed to transfer water from low-valued uses to high-valued uses during a period of intense scarcity. Transfers were needed because the water allocations during non-drought periods did not necessarily reflect the high-valued uses during a drought. The Banks were very effective at achieving this because they minimized transaction costs and risks. Thus, it can be said that if transaction costs and risks can be minimized by either the implimentation of new government policies or the abolishment of old government policies, the water market can operate more efficiently. One can take this further by saying that water would be allocated more efficiently if water markets were allowed to operate freely. That is, water would go to high-valued uses in the absence of government regulations and laws that limit water transfers and raise transactions costs. Specifically, riparian rights, which are currently non-transferable in California and elsewhere, should be made transferable. They were transferable while the Banks were in existence, and were a major factor in the success of the Banks. If property rights were clear, secure, and transferable, owners of the water (or the right to the water) would allocate it in such a way as to maximize the present value of net benefits. This would ensure the continuity of the water supply except in the presence of an extremely large discount rate. Some would argue that the success of the Banks indicates a need for more government intervention. That is a superficial conclusion, however, because the Banks were, in a sense, temporary exceptions to policies that limit transfers.

If the market were allowed to operate more freely, however, some legislation would still be required in order to limit externalities. In-stream uses which are often undervalued would need to be taken into account. In addition, environmetal impacts that do not directly effect the owner of the water must be accounted for with government policy. The combination of market forces and properly targeted policies would lead to an efficient water allocation which would be sustainable both in terms of the water supply and the overall environment.

Source: Peter Bingenheimer, Colby College ’96