Investigations On Interaction Between Electricity Markets And Emissions Trading Markets

Climate change has become an important and widespread concern of the public around the globe. How to mitigate the ever-worsen climate change problem? Introducing the emissions trading (ET) is an effective way to curtail the emissions of greenhouse gases. ET has been implemented in the European Union, and has been proposed in a number of other countries including Australia and the United States. The power industry is the largest emitter of greenhouse gases in many countries of the world, and should implement the ET into it to improve the status of power sector emissions. However, the emissions trading will inevitably have some impacts on power industries and electricity markets. These potential impacts include some key issues associated with the Emissions Trading Scheme (ETS) design, the types of emission permits trading, as well as the methods of allowance allocations.The objective of this dissertation is to examine the interactions between ET and generation companies (GenCos) involved with emissions, such as coal-fired and gas-fired companies, and those not involved with emissions, such as wind and nuclear power generation companies. The short-term and long-term impacts of emissions trading on the behavior of GenCos are studied. The short-term impacts include the bidding strategies of GenCos and the markets equilibrium of gaming among GenCos participating in both the ET market and electricity market. The long-term impact is mainly concerned with the changes of investment strategies after the implementation of the ET.Specifically, this dissertation is comprised of five chapters.1. The emerging electricity market is more akin to an oligopoly market. A GenCo should analyze the price fluctuation of emission rights as well as that of electricity, and the response of its rivals so as to maximize its own profit. In the first chapter, a model used to forecast the price of emission permits of carbon dioxide is established and a Supply Function Equilibrium Model is derived to analyze the action of GenCos who gamed in both the emission trading market and electricity market. Simulation results show the implementation of emission trading could limit the emission but lead to the electricity price increase and the―windfall profit‖of some GenCos.2. The second chapter extends the type of emissions permits. Not only the permits of carbon dioxide but also those of sulfur dioxide and nitrogen oxide are traded in the emissions trading market. By using the well-established conjectural variation method, a mathematical model is developed for investigating the strategic behavior of GenCos participating in both the ET market and the electricity market. Four scenarios with GenCos playing games of Cournot, Cartel, Bertrand and Stackelberg are demonstrated respectively. The strategic behavior in each scenario is examined. As expected, the results show that the implementation of the emissions trading will increase the electricity market clearing price (MCP), limit the emissions form high-emission GenCos, lead to the outputs from those GenCos with low-emission increased, and then curtail the overall emissions from GenCos.3. In a bilateral contract market, the GenCos could trade with customers directly. In the third chapter, the complementarity method is used to simulate the equilibrium among customers, fossil-fueled generation units, wind power units and the grid company, which participate in the emissions trading market and the day-ahead electricity market. Forward contracts, the operating reserve market, the fluctuation of generation outputs from wind GenCos and the price of emissions allowances (EA) are considered in this model. A program of General Algebraic Modeling System (GAMS) is developed for simulations. Simulation results show that the ET could increase the share of the generation output of wind GenCos, and decrease the emissions from fuel-fossil GenCos; the bilateral contracts between GenCos and users could limit the ability of GenCos’exercising market power by driving up electricity price; when the emissions allowances allocated decrease, the price of EA will increase, and hence the dispatch orders of different generation units will changed. Because of the cost of wind GenCos is still very high, it will be more realistic to increase the market share of wind GenCos by reducing cost rather than by relying on the ET implementation.4. In the fourth chapter, a mathematical model is developed for investigating the equilibrium state of an interperiod multi-market with different kinds of market participants playing in energy, reserve, point-to-point financial transmission right (FTR) as well as emissions trading markets. This complicated problem is formulated as a well-established mixed linear complementarity problem (MLCP). The market participants considered include generation companies, FTR owners, arbitrageurs and electricity consumers. A GAMS program is coded to simulate the MLCP. Finally, the results are served for demonstrating the essential characteristics of the developed model and method. This is a useful exploration to analyze the Nash equilibrium state of multiple players participating in multiple markets.5. Uncertainties of generation investment increase after the implementation of the emissions trading, and the traditional net present value analysis is no longer able to meet the needs of power generation investment decision-making. In the fifth chapter, in the framework of the real options theory, some uncertain factors are integrated into a model to help the investors to make decision, including fluctuations of the electricity price, fuel price and emission rights price, policy change, carbon capture and storage technology development. Investors could choose to invest in conventional coal-fired power, CCGT units, nuclear power, wind power and other renewable energy. First, we assumed that the variation of the uncertainties follow Geometric Brownian Motion (GBM). Secondly, the software of EViews 6.0 is used to develop a statistical relation from the history of time series, such as the price of electricity and permits of emissions. An Generalized Autoregressive Conditional Heteroskedasticity (GARCH) model, GARCH(1,1), is applied to obtain the parameters for prediction of each uncertainty. Finally, a real option model is established to evaluate each investment project, and the software named Crystal Ball is utilized for the Monte Carlo simulation in order to help the investor make decision in choosing the best project and the best time of investment. The effectiveness of this model is verified by simulations.