Essays on Uncertainty and Decision-Making in Environmental Economics

This draft PhD thesis analyzes four issues in the research field that consists of the intersection of environmental economics, uncertainty, and decision-making. 1. Economic investments that lead to environmental damages beyond repair - like cutting rainforests or emitting pollutants with long life times - can be interpreted as irreversible investments with uncertain pay-offs. In the context of irreversible investments under uncertainty two different concepts of option value were developed. They account for the value of keeping options open - i.e. postponing the irreversible investment - in the presence of potential future learning about the social profitability of the investment. These are the concept of quasi-option value (value of information) developed in the environmental and resource economic literature and the concept of real option value (value of the option to postpone) developed in the financial economics literature. This thesis shows that the quasi-option value (value of information) is that part of the real option value (value of the option to postpone) that is related to uncertainty being resolved over time. This means that common knowledge in the environmental and resource economic literature with respect to the relation of both concepts is not correct. 2. If marginal damages of pollution rise with the pollution level, low marginal pollution abatement costs demand a relatively low Pigouvian tax - the pollution tax level that leads to optimal abatement. As a result, to determine the Pigouvian tax level, a regulator needs to know pollution abatement costs, information that is privately owned by individual pollutors. Unfortunately, pollutors have no intention to reveal their true abatement costs. One text book solution to this problem is that the regulator keeps adapting the tax level until the pollution tax leads to an emission level resulting in marginal damages equal to the tax level - the criterium for social optimality. That is, only after an unspecified amount of time and abatement adaptation costs the socially optimal abatement level is reached. This thesis shows there exists a non-linear tax system, the so called differential tax that triggers firms to reveal their abatement costs truthfully and will lead to socially optimal abatement levels instantly, thereby reducing social costs compared to the iterative solution. 3. In the presence of environmental innovation, future abatement costs are unknown by the polluting firms as well. Under those circumstances the regulator is needs a tax system that is flexible with respect to falling abatement costs through innovation (please recall that lower marginal abatement costs demand a lower Pigouvian tax) as well as keeps incentives for innovation - created by pollution taxes after innovation - upright. This thesis analyzes the effects of a robust second-best pollution taxation scheme that combines flexibility to cope with changing marginal abatement costs with commitment to relatively high tax levels to create incentives for environmental innovation. It shows that this instrument is superior to time consistent taxation. Superiority of the flexible taxation scheme with respect to commitment to a linear tax could not be shown. 4. The use of quantitative model results to support policy making is one way social sciences might have a positive effect on real world decision making. This thesis identifies decision-maker related potential errors made in the interpretation of quantitative model results. It is argued these errors can be prevented by the social scientist by stating explicitly for which policy maker and for what purpose the model results are meant for