Irrigated Agriculture under Stress: Discussion

Abstract

The paper by Robie is primarily a series of observations of legal, economic, and institutional problems highlighted by a recent twoyear drought. Apparently, the abilities to adjust to lower water supplies exceed the previous expectation of most people. This result should not have been surprising since much of the arid world survives very well using much less agricultural and municipal water than we do. Any thoughtful economist should have anticipated that water use will be high when it is cheap and plentiful and that, even in the short run, adjustments can be made to raise the average productivity of water as it becomes more scarce. Robie expressed surprise at the willingness of farmers to pay more than $40 per acre-foot for water. However, in a longer run setting, we should expect such a phenomenon to be common with concomitant adjustments in other resource values, particularly that of land. As time passes and the opportunities and incentives for adopting different technologies and management practices are presented, we should expect the ability to survive with less water to improve significantly. What we have seen are normal adjustments to a resource having greater relative scarcity. On further thought, we should not be so surprised by a drought of this length or severity when our experiences into the past are so brief. Archeological research collecting weather information from the past 5,000 years has shown that equally or more severe droughts have persisted for periods of twenty to forty years. Thus, scientists, including economists, should expand their thoughts and research to consider the necessary preparations for and potential adjustments to such a phenomenon. We might take heed of Kelso's comments in the fellow's lecture and make plans that do not require using all available resources all the time. The paper by Burt, Cummings, and McFarland provides an analytical construct for estimating optimal rates of groundwater mining. The authors use linear programming to generate data for a function that in turn estimates income for alternative levels of capital stock and groundwater withdrawal rates. They show that, as the discount rate goes up, there is a greater willingness to mine the groundwater and capture its value in the present rather than preserve the water for future use. The higher the discount rate, the more mining that is allowed before the desired steady-state level of groundwater stocks is achieved. While this is a neatly packaged device for generating such information, the result is exactly what one should expect. So, while the Burt, Cummings, and McFarland procedure or others could be used to provide this kind of information, there remains the important question of how it can be put to use. It is not obvious how local or state policy makers are going to be led to the optimum discount rate or rates of declines in the static water level. Those who ultimately choose discount rates and, in turn, the rate of groundwater mining are often politically elected or appointed. This leads to a very short planning horizon and implicitly imposes a high discount rate on the decision process. The result too often is to exhaust the resource in a relatively short time period in order to satisfy the short-term demands of political constituents. The opportunities for using the water are eliminated for the future when societal needs for potable or fresh water may be much greater than at the present. Kelso's reversability criteria are violated. Some factors that need consideration in a water-mining situation but not adequately discussed in this paper include: capital assets should not require more rapid depreciation due to declining water levels than their normal wear-out period; the impact of rising energy costs or energy shortages anticipated for the future should be considered; and, importantly, the application of such information cannot be