Modeling the Hedging Decisions in Electricity Markets Using Two-stage Games

The electricity market structure aims to provide effective competition among producers. However, the exercise of market power considerably affects energy prices and induces the wealth transfer from consumers to generators. It is widely recognized that the incentive of generator to exercise market power depends on its hedging decision. The hedged generator has very little or no tendency to exercise market power. However, existing approaches for forecasting market power typically take the hedge cover level as an external input to the model, providing an open-loop analysis. In practice, the level of hedge is chosen by a generating firm and treated confidentially. Omitting this assumption results in equilibrium constraints in the forward and spot stages of the trade period. This type of problem can be modelled as a two-stage game, where in the first stage the generator decides on the profit-maximizing level of hedging and in the second stage it takes part in Cournot game on output in spot market. We show that in the oligopoly case with no constraints on installed capacity and transmission limits there is no equilibrium in pure strategies, which corresponds to the "all-or-nothing" decision. In the case when transmission and capacity constraints are binding, we use a backward induction approach and closed-loop analysis to find a sub-game perfect Nash equilibrium (SPNE) on the level of hedging and a corresponding SPNE in spot market. We show that the solution in this constrained network is similar to the unconstrained one.