Problems of uncertainty, learning, and welfare measurement in resource and environmental economics
This dissertation is a collection of three essays that focus on problems concerning uncertainty, learning, and welfare measurement in resource and environmental economics. Each essay focuses on a separate aspect of these problems. The first essay analyzes search, learning, and economic performance in a dynamic fishing game played by independent fishermen and by members of a stylized fishing cooperative. This paper extends our understanding of the behavior of fishermen operating in a fishing cooperative and highlights the importance of the design of fishing policy to promote their use. Devising contracts that result in optimal investment in information may be particularly challenging in fisheries, due to the club good characteristic of information, its costly acquisition, and the common property nature of the fishery resource. Contracts commonly used to prevent effort-shirking introduce a free-riding problem on the costly information investments of other members. The second essay presents a dynamic transition model to study the economic fundamentals that determine the path from an initially over-capitalized fishery to the long run cost-efficient fleet structure after the introduction of an individual transferable quota (ITQ) program. The results suggest that beliefs play a strong role on the transition. Uncertainty by fishermen over their relative cost-efficiency translates into uncertainty over quota trading prices. As such, a component of the ITQ asset's market price is speculation over future trading prices. Heterogeneous uncertainty and learning also provide insight into the observed pattern of vessel exit, generating new insight into the slow transition observed in U.S. fisheries. These results should be of interest to policy makers. The near universal practice of allocating the initial endowment of quota based on historic catch promotes delayed-exit strategies on cost-inefficient vessels and thereby prolongs the transition period during which the fully efficiency benefits of ITQ management are unrealized. The third essay uses the cleanup of Luke Air Force Base (a deleted Superfund site in Maricopa County, AZ) as a quasi-experiment to perform inside sample and outside sample validation exercises on the pure-characteristics vertical sorting model. Based on these validation exercises, the average bias (absolute deviation) on housing expenditures is approximately 20% and nearly 25% for income. In this setting, the model appears to predict the movement of population shares, but tends to under-predict the housing prices in response to remediation of the Superfund site. The spatial distribution of welfare is also a feature of the structure of the vertical sorting model.