This dissertation presents three studies on the economics of water use and quality. The first two chapters analyze farmers' irrigation behavior both empirically and theoretically. Finally, the third chapter looks at how effective regulations are in changing firms' water polluting behavior.In chapter 1, I analyze how climate variability influences farmers' irrigation behavior. The sustainability of irrigated agriculture has come into question due to the declining fresh water supply and ever-increasing demand for water from other sectors due to population growth. Uncertainty brought upon by ongoing climate change further complicates the situation. An increasing number of studies now look at the potential impacts of climate change on irrigation water use via simulations based on crop and hydrology models. While irrigation strategy can be specified in their models, it entails limited flexibility, forcing farmers to behave rather mechanically, following a set of rules pre-specified by the researchers. Econometric approaches, on the other hand, allow us to elicit the functional relationship between climatic conditions and irrigation water use in which farmers' behavioral aspects are embedded implicitly. Previous econometric studies have greatly simplified the relationship between irrigation water use and climate by simply linking aggregate climate variables to seasonal irrigation amounts. In this study, we advance the econometric approach by developing an irrigation water use model based on the agronomy and irrigation science literature. Further, we employ an econometric specification to capture farmers' potentially nonlinear response to daily as opposed to annual climate variables. Econometric analysis on well-level groundwater extraction data in Nebraska shows that farmers indeed respond nonlinearly to daily climate variables and thus that the use of time-aggregate climate variables is not desirable. Further, by comparing regression results on the real data to synthesized data generated by AquaCrop, a crop simulation model, we find that actual farmers do not respond as strongly to climate variables as would a hypothetical farmer programmed to follow a soil moisture target strategy. This indicates that previous predictions of changes in irrigation water use due to climate change may have been overstated.In chapter 2, I examine the implications of energy supply interruption on groundwater consumption for irrigation. In groundwater-irrigated agriculture, energy and water uses are inextricably linked as energy is required to pump water from the aquifer. The direct link between these resources implies that a policy change on one side influences the other side as well. Energy suppliers commonly use a irrigation load control program in which energy supply may be cut off in order to manage peak energy load. Under energy supply interruption, farmers need to take into account the possibility of random input supply interruption when deciding the timing and amount of irrigation. I develop and solve a stochastic dynamic optimization problem of irrigation schedule under random input supply interruption, in order to understand how farmers adapt their irrigation strategy and the consequent effects on seasonal groundwater consumption. I find that farmers will increase the amount of irrigation per irrigation opportunity to hedge against the risk of not being able to irrigate in future periods. The impact of energy supply interruption on seasonal groundwater consumption, however, cannot be signed analytically and is an empirical question. Numerical examples with a model calibrated to corn production in Nebraska show that the total amount of irrigation may indeed go up as a consequence of a load control program in the region. This indicates that a policy that is effective for energy demand management could provide agricultural producers with a perverse incentive in terms of groundwater consumption. As a result, joint energy-water management planning in groundwater-irrigated agriculture can be beneficial. Finally, in chapter 3, I examine the deterrence effects of regulatory activities on point-source pollution discharges into bodies of water under the National Pollutant Discharge Elimination System (NPDES) program established as part of the Clean Water Act (CWA). The contribution of this study lies in advancements in econometric strategy that allow a particularly difficult endogeneity problem to finally be addressed. Estimating the true impacts of regulatory activities on polluters' environmental performance is inherently difficult due to a reverse causality: firms that emit more attract more regulatory actions. The fixed effects estimation method has been commonly used as it avoids this reverse causality by limiting identification information to variation over time, discarding cross-sectional variation. We show that the strict exogeneity assumption, necessary for the consistency of fixed effects estimators, can be violated due to the dynamic interactions between regulators and polluters. I employ a generalized method moment estimation using moment conditions based on sequential exogeneity assumptions that are weaker than strict exogeneity. The fixed effects estimation results indicate that non-monetary sanctions would increase the amount of Total Suspended Solids (TSS) discharged and that fines have no impact on reducing TSS discharges. On the contrary, GMM results indicate that non-monetary sanctions have no impact but fines have significant deterrence effects.
PDF
download
878987797
028-85220240
