Modeling the unilateral multi-part strategic withholding in electricity markets

Modeling and predicting the exercise of market power is an important problem for modern electricity markets. A change in the generation mix and an evolvement of bidding procedures give ascent to a variety of new types of strategic behavior involving elaborate strategies with unconventional strategic variables. In this paper we show how it is important to consider all parameters of the bid to perform a sound market power assessment. We propose a bi-level model of a profit-maximizing decision maker bidding to the electricity market. A company, owning a portfolio of generating units, can strategically choose a price offer, capacity and a ramp rate of its units. We show that withholding the flexibility of the sources causes considerable market power costs. The strategic decision of the company is followed by the optimal dispatch, where the market price is calculated. The problem is formulated as a mathematical program with equilibrium constraints (MPEC) and recast as a single-stage mixed-integer linear programming problem (MILP). Developed model is applied to a 24-node IEEE test system to highlight the withholding and demonstrate the cost of market power.