Structural Estimation of Switching Costs for Peaking Power Plants

This paper estimates costs associated with mothballing, restarting, abandoning and maintaining peaking power plants. The paper develops a real options model to explain switching and maintenance behavior of plant managers. The constrained optimization approach to estimate crucial costs accommodates non-parametric dynamics for the expectations of the plant managers regarding future profitability. The empirical analysis is based on a database of the annually reported status of power plants to the United States Energy Information Administration (EIA) during 2001-2009. We arrive at economically meaningful estimates of maintenance costs and switching costs, and discuss these in light of rates used in the Pennsylvania-New Jersey-Maryland capacity market.