Natural gas has replaced coal as the dominant fuel for US electricity generation. However, utilities in regulated US states have retired coal more slowly than others. We build a structural model of rate-of-return regulation during an energy transition where utilities face trade-offs between lowering costs and maintaining and using legacy capacity. A regulated utility facing carbon taxes lowers short-run coal generation 48 percent as much as a cost minimizer would. Thirty years after a sudden energy transition, a cost minimizer has retired 71 percent more coal capacity than the regulated utility. Alternative regulations may jeopardize affordability and reliability goals during energy transitions.

We provide empirical evidence that automobile manufacturers use cash incentives to offset how gasoline price fluctuations affect the expected fuel expenses of automobile buyers. Regressions based on a database of incentives over 2003 to 2006 suggest that on average, manufacturers offset 40% of the change in relative fuel costs between vehicles due to gasoline price fluctuations. The results highlight that carbon taxes and emissions trading programs likely would generate substantial substitution within vehicle classes, and studies that ignore manufacturer discounting likely underestimate consumer demand for fuel economy. The results also have implications for the optimal design of feebate programs.

The US Environmental Protection Agency (EPA) uses a dynamic approach to enforcing air pollution regulations, with repeat offenders subject to high fines and designation as high priority violators (HPV). We estimate the value of dynamic enforcement by developing and estimating a dynamic model of a plant and regulator, where plants decide when to invest in pollution abatement technologies. We use a fixed grid approach to estimate random coefficient specifications. Investment, fines, and HPV designation are costly to most plants. Eliminating dynamic enforcement would raise pollution damages by 164 percent with constant fines or raise fines by 519 percent with constant pollution damages.

Government policy uncertainty affects irreversible decisions including technology adoption and exit. This paper quantifies uncertainty surrounding the Mercury and Air Toxics Standards (MATS). We estimate a dynamic oligopoly model for coal-fired electricity generators that recovers generators’ beliefs regarding future MATS enforcement. We develop the approximate belief oligopoly equilibrium concept, where players understand that their decisions impact aggregate market states. MATS enforcement created substantial uncertainty: the perceived enforcement probability dropped to 43%. Resolving uncertainty early would increase profits by $1.39 billion but also increase pollution costs by 0.652–1.776 billion. Had exit been unlikely, resolving uncertainty early would have decreased pollution.