The research project is motivated by problems in environmental economics concerning the use of market mechanisms to control green-house gas emissions. It concentrates on the design and the analysis of cap and trade schemes, and uses as starting point, the failure of the first phases of the California RECLAIM program and the European Union ETS to reduce emissions. Some of the technical issues raised by this research involve new mathematical models in environmental economics, very general competitive equilibrium problems, large scale stochastic control problems, and ill-posed inverse problems. They all lead to new mathematical challenges in optimization, especially in infinite dimensional spaces, parameter sensitivity analysis and robustness, and stochastic numerics, including multi-scale asymptotic approximations and Monte Carlo simulations. The work will contribute to the important role that mathematics has to play in major policy making decisions.
The research that will be carried out with this award is designed to empower policy makers in designing emissions markets capable to meet emissions target while at the same time, reducing the overall social costs as well as the obscene windfall profits energy producers can make when markets are poorly designed. In the area of economics of global warming, where most of the models and large scale simulation programs are deterministic, there is still a strong resistance to the introduction of probabilistic thinking. Poor regulation and a lack of understanding of the stochasticity of the risks involved, have contributed to embarrassing public policy faux pas in previous emissions market designs. By developing new tools and strategies within the realm of stochastic analysis, and by showing how they can be brought to bear on the regulation of emissions markets, the results of this research will help educate the future scientists about what financial markets can and cannot do, and policy makers and regulators about what equilibrium models can explain and justify. By focusing on stochastic models which have traditionally remained off the radar screen of practitioners and policy makers, it will raise the level of awareness for the power of probabilistic modeling in a wide range of domains.
