Degree Type


Date of Award


Degree Name

Doctor of Philosophy



First Advisor

Todd Sandler


The main issues of this dissertation are those problems that are associated with the control of stock externality under the constraint of the rigidity of policy variables. An optimal control on a stock externality requires for the control variables to be adjusted every moment of decision, which is not available in the real world. For the description of such an institutional constraint, a fixed one-time control policy has been introduced. The justification of this assumption is that, under a positive rate of discount, there will not be any great difference in the controller's current decision making whether a policy variable can be changed far in the future or completely fixed forever;Chapter 1 gives a brief introduction of the study in the subsequent Chapters. In Chapter 2, a general review on the theory of externality is provided by examining the concept, the sources and the possible remedies of externality. Based on this review, the directions of future studies in this area are suggested;The main point of Chapter 3 is to clarify the effect of the rigidity of a policy variable. One of the important results is that, if the initial state of the environment is more (less) deteriorated than the first-best optimal state, then the steady state of the stock under the second-best control will end up with less (more) deteriorated than the first-best steady state which is independent of the given initial state;Chapter 4 examines the issue of optimal starting point of fixed one-time control. First, a general rule has been formulated for the decision of optimal starting point. And then, it is proved that there can be an efficiency gain by delaying the start of the control unless the current state of the environmental stock is more deteriorated than its first-best steady state. It is shown that higher levels of the administrative costs may provide the controller an incentive for an unduly long delay in the control;The issue in Chapter 5 is the relative performance of two representative control modes, a Pigouvian tax system and a quantity restriction. The result is that the random factor in the benefit function is the only relevant source of uncertainty that affects the relative economic performance of the two policy schemes. However, this random factor does not reverse the ranking of the two policies, but only affects the magnitude of the difference of the welfare states. A high discount rate and a high amelioration rate turn out to be important factors that make the tax system more attractive and vice versa.



Digital Repository @ Iowa State University,

Copyright Owner

Il-Dong Ko



Proquest ID


File Format


File Size

105 pages