Groupe d’études et de recherche en analyse des décisions

Equilibrium capacity expansion under stochastic demand growth

Alfredo Garcia University of Virginia, États-Unis

In critical energy infrastructure sectors (e.g. electric power generation, natural gas transportation, oil refining capacity) maintaining certain level of excess capacity is socially valuable (as it serves to protect against unexpected market conditions) but not necessarily compatible with the incentives for individual firms in the market. In this paper, we develop a dynamic oligopoly model with a stochastically growing demand to analyze the inherent tension in market-based incentives for capacity expansion where capacity additions take place over long time lags. Our results indicate that the market fails to induce the socially optimal level of capacity. However, the magnitude of this failure varies greatly as a function of entry costs and the relative profitability of investments in the market (as measured by the ratio of maximum markup over production costs and investment costs). In general, the likelihood of insufficient capacity in equilibrium increases with decreasing probability of demand growth, increasing discount and depreciation rates and/or increasing investment and/or production costs. We discuss the public policy implications of our results including the implementation of capacity markets.

Keywords: Capacity expansion, critical infrastructure, dynamic games, Markov perfect equilibrium.