The Economics of Wilderness Preservation and Timber Supply: Problems in Dynamic Optimization and Uncertainty.

This thesis considers three problems related to the harvest of timber and preservation of wilderness on public l and s. The first problem concerns the comparative statics properties of a model of timber production in a steady-state and their use in characterizing the competitive supply curve for timber products. It is found that when initial silvicultural efforts is a choice variable, comparative statics results are complex and ambiguous. When a plausible restriction is placed on the biological growth function and entry of new firms is considered, an upward-sloping supply curve is expected. The second problem is the determination of the optimal harvest of old-growth forests and transition to a steady-state in a model where old-growth forests have value as wilderness. Using an optimal control approach, it is established that the optimal steady-state wilderness preserve is a decreasing function of the cost of production in second-growth st and s and an increasing function of the cost of old-growth harvest and of the interest rate. The efficient level of silvicultural effort in managed st and s is characterized and it is shown that the optimal wilderness preserve is largest when the efficient level of effort is applied. The prospect of technological change for second-growth st and s is shown to increase the optimal wilderness preserve. The problem of small initial stocks of old-growth is addressed. The third problem concerns the treatment of uncertainty and learning in models of development of natural environments. The effects of incorporating learning into the model relative to when learning opportunities are ignored are studied. Two approaches are contrasted: the effect on expected net benefits and the effect on optimal investment decision rules. Linear and non-linear models are analyzed, as are perfect and imperfect information. It is shown that initial investment decisions are more conservative when learning possibilities exist and that the "shadow tax" on initial development which reflects anticipated learning increases with increases in ex ante risk.Ph.D.EconomicsUniversity of Michiganhttp://deepblue.lib.umich.edu/bitstream/2027.42/160322/1/8502826.pd