| This dissertation is comprised of three papers in two distinct areas in environmental economics and policy. Chapter 1 examines questions of environmental justice: specifically, it analyzes the relationship between a polluting firm's propensity to violate environmental regulations and the socio-economic composition of the community in which that facility is located. A theoretical model is developed which generates empirically testable predictions about this relationship, predictions which differ depending on whether the regulator is minimizing social costs or engaging in discriminatory behavior. Empirical testing of this model using data on firms regulated under the U.S. Clean Air Act finds that firms are more likely to violate environmental regulations if they are located in communities with a high fraction of minority residents, thus lending some support to the hypothesis of discriminatory behavior by environmental regulators. Chapters 2 and 3 analyze different aspects of the so-called "judgment proof problem," which arises where a regime of strict liability for damages from environmental accidents results in liabilities that exceed the value of firms' financial assets. Chapter 2 examines in detail the extent to which a policy of extending liability to third parties with whom firms interact can mitigate the judgment proof problem. It is found that where firms can engage in different types of care activities to reduce expected damages from environmental accidents---and where some of these care activities are observable to the regulator while some are not---extended liability can, in fact, be welfare improving. Chapter 3 turns to the question of the appropriate way in which to model firms for whom the judgment proof problem is at interest in an environmental context. This chapter makes clear the implicit assumptions inherent in many of the existing modeling approaches to date, and identifies certain problems that arise when these assumptions are made explicit. A new approach in which firms are modeled as infinitely-lived dynamic optimizers---is shown to avoid these problems, and is thus suggested as more appropriate means by which to analyze the judgment proof problem. |
