Robustness of capacity markets: a stochastic dynamic capacity investment model

We present a stochastic dynamic capacity investment model that incorporates strategic bidding of generators, price elasticity of demand, and ramping constraints. The model is based on an iterative procedure of finding an equilibrium electricity price forecast closely depicting real-world investor behavior. We assess the robustness and the total cost of generation of a capacity market compared to an energy-only market as capacity markets are increasingly considered worldwide due to market imperfections and rising feed-in from renewables. To estimate unplanned unavailabilities, we analyze a novel data set of the European Energy Exchange describing unplanned unavailability incidents. Applying the model to the Great Britain market, we find that capacity markets are more robust to unplanned unavailabilities of conventional generation in terms of the total cost of generation and price volatility. At the same time, capacity markets induce additional costs through capacity payments. Therefore, we jointly analyze the total cost of generation and the robustness with different levels of reserve margins. Our results show that there is a promising area of efficient trade-offs between costs and robustness in the range of reserve margins between 0 and 15%.