The optimal pricing of natural resources

The first two essays of this dissertation examine how farmland prices can be used to estimate the impact of climate change on agriculture. The first study presents a hedonic model of farmland value and shows how the omission of irrigation in previous studies has biased earlier results, as well as how different weighting schemes might exacerbate this bias. Even though some of the previous studies predicted that agriculture may benefit from climate change, the calculation of confidence intervals reveals that these earlier point estimates were not significantly different from zero.;The second essay presents revised estimates where the impacts of climate change are derived separately for dryland non-urban, and irrigated and/or urban areas. Special emphasis is given to the spatial correlation of the error terms. When the analysis is limited to dryland and non-urban counties, the different estimators converge and the confidence intervals are cut by up to half. Dryland U.S. agriculture is unambiguously damaged under the CO2 doubling scenario, and the damages are significant and quite large relative to recent estimates in the literature. Irrigated agriculture will face increased water shortages, but the aggregate costs are smaller than for dryland agriculture.;The third essay presents a model for the optimal extraction and exploration of an exhaustible resource with an uncertain stock size. It is shown that exploration should be started as soon as the stock of known reserves falls below a critical level. This critical level is decreasing in the unexplored area. When exploration is conducted by competitive firms, the price of reserves fluctuates below a constant maximum attained along the critical level until all unexplored area is exhausted. Nevertheless, expected price always rises at the rate of interest, as implied by Hotelling's rule. However, time series samples seem to, with high probability, indicate that the path of conditional price expectations is rising at a rate below the rate of interest. Numerical simulations confirm that realized prices are likely to follow a saw-tooth pattern with a common constant finite (local) maximum until unexplored resources are exhausted.