Three empirical essays on power plant operating decisions

Abstract

Switching from coal-fired to natural gas-fired generation and increasing the thermal efficiency of fossil fuel fired generation plants have been identified as ways of achieving meaningful emission reductions. In chapter 2, we examine the fuel-price responsiveness across gas plant technologies and across the market structures in which the plants operate. We find that there are significant differences in the generation and efficiency responses of gas plants to fuel prices across generation technologies and market structures. Using the parameter estimates from our econometric models, we calculate emissions savings from efficiency improvements and fuel-switching possibilities. The large and local economic and environmental footprint of power generation makes a salient target for state politicians when lobbying the electorate. Researchers have utilized the fact that many states have term limits for governors to determine how changes in electoral incentives alter state regulatory agency behavior. Chapter 3 asks whether the change in electoral incentives, and its impact on regulatory agencies, spillover into private sector decision-making? We find strong evidence that power plants invest less in water pollution abatement if the governor of the state where the plant is located is a democrat and term-limited. Furthermore, the difference in plant decision-making is strongest when the governor wins their term-limited term by a narrow margin. Finally, we show that the lack of investment has environmental impacts by increasing thermal pollution and chlorine use by the plant. Natural gas power plants can further specify their procurement contracts with pipeline distributors using a{\it firm} contract option that guarantees delivery at an additional cost. Chapter 4 uses transaction level data to empirically test what characteristics lead to use of{\it firm} contracts and how the premium for firm contracts changes with these characteristics. We find that smaller plants, plants located in states with more variance in electricity demand, and plants in states with more inflow pipeline capacity are statistically less likely to use a{\it firm} contract.{\it Firm} contracts are on average 2.5\% more expensive and this premium increases as the weather is colder and the state a plant is located in has less inflow capacity.