"Firm Behavior in Pollution Permit Markets.".

University of California, Berkeley. Outstanding Ph.D. Dissertation Award.

Policy makers are increasingly relying on emissions trading programs to address environmental problems caused by air pollution. These programs can significantly affect product market outcomes in polluting industries through their effects on firms' production and investment incentives. Conversely, the structure and economic regulation of polluting industries can undermine the efficiency of pollution permit markets. These inter-market interactions raise a series of interesting questions that motivate the three papers in this dissertation.

If polluting firms in an emission trading program face different economic regulations and investment incentives in their respective industries, emissions markets may fail to minimize the total cost of achieving pollution reductions. In the first chapter, I analyze an emissions trading program that was introduced to reduce smog-causing pollution from large stationary sources (primarily electricity generators) in 19 eastern states. I develop and estimate a random-coefficients discrete choice model of a plant's environmental compliance decision. Using variation in state-level electricity industry restructuring activity, I identify the effect of economic regulation on pollution permit market outcomes. There are two important findings. First, plants in states that have restructured electricity markets are less likely to adopt more capital-intensive compliance options. Second, this economic regulation effect, together with a failure of the permit market to account for spatial variation in marginal damages from pollution, have resulted in increased health damages. Had permits been defined in terms of units of damages instead of units of emissions, more of the mandated emissions reductions would have occurred in restructured electricity markets, thereby avoiding on the order of hundreds of premature deaths per year.

The second chapter focuses on short-run compliance decisions made by firms in an emissions trading program. According to the Coase theorem, if property rights to pollute are clearly established and emissions markets nearly eliminate transaction costs, the market equilibrium will be independent of how the permits are initially allocated across firms. This important hypothesis has been difficult to directly test on account of the likely endogeneity of facility-level permit allocations with respect to emissions. Using panel data from Southern California's RECLAIM program, we evaluate the independence of equilibrium firm-level emissions and permit allocations.

The third chapter looks at Cap and Trade (CAT) regulation as it applies to global climate change. For political, jurisdictional, and technical reasons, CAT programs designed to address global climate change regulate only a subset of polluting firms. When participation is incomplete, pollution reductions among program participants can be substantially offset, or even eliminated, by "leakage" of production and emissions to unregulated producers. I develop a model of an oligopolistic industry in which only a subset of producers are subject to CAT regulation. The model accommodates strategic behavior in the product market and vertical arrangements between wholesalers and retailers.

I find that the existence of a forward product market exacerbates leakage problems. Simulation results suggest that updating firm-level permit allocations based on past production can mitigate leakage problems, but not eliminate them.