Mitigating Strategic Bidding in Transmission-constrained Electricity Markets

Abstract

In the last few decades, the European Union has directed most of its efforts towards the creation of an internal market, free of internal borders, where competition reigns and the invisible hand guides the market to a social optimum. The assumptions on which this ideal is based, however, do not always hold in practice, especially when it concerns a good as peculiar as electricity, which sets itself apart from other commodities through a number of its physical characteristics. In their attempt to safeguard their strategic objectives of acceptability, accessibility and affordability of power for end users, policy makers face two strategic challenges, arising largely from the physical characteristics of electricity: market power and Congestion. Congestion may be dealt with in a number of ways, but congestion management schemes strongly impact the dynamic interactions in the market, possibly exacerbating market power. Given the high cost associated with ”getting it wrong”, a thorough understanding of the workings of these mechanisms is required. This thesis aims to increase the understanding of Financial Transmission Rights, a risk hedging tool for market parties in electricity markets with inter-zonal price uncertainty. More specifically, the work herein attempts to better understand the strategic bidding options producers have, and the effectiveness of market regimes designed to counteract them. By measuring the welfare allocation and price distortion that result from strategic producer bids, the policies can be contrasted to a no-policy scenario, and their effectiveness may be assessed. The methodology used in this thesis is that of a multi-level optimisation procedure, based on supply function equilibria for generation portfolios consisting of multiple generators. Producer action spaces are discretised using multipliers based on the competitive scenarios. Once the action spaces have been defined, each action combination in the upper level problem (the market for FTRs) results in a static non-competitive Nash equilibrium of the lower level problem (the power market), which is obtained by solving the market for all possible action combinations given the FTR strategy, and determining the equilibrium strategies. Using these results to populate the pay-off matrix for the upper level problem, the Nash equilibrium in this upper-level problem determines the outcome for the given parameter space. The interpretation of results is limited by the assumptions of the market being a duopoly and fully static, absence of price elasticity of demand, complete information, and imperfect information regarding the true cost functions of generators. These assumptions correspond to what might be termed a worst-case scenario for policy makers. The conceptual model is translated into a Python-based simulation environment. This environment is then used to determine effectiveness of the following policies in a network consisting of three nodes, four generators and one load-serving entity: a hybrid transmission right, largely corresponding to an FTR with a use-it-or-lose-it clause, and a hedging-only requirement imposed on the FTR portfolios producers hold. The former policy aims at counteracting the leverage effect from losses in one market on profits in the other, while the latter focuses purely on countering sign changes of nodal price differences. Analysis leads to the following insights: The strong interdependency between FTR and power markets is once more shown: producers are able to leverage losses in the power market against greater gains in the FTR market. Evidence is found for producers being able to leverage their FTR profits by taking small losses in the power market. The Hybrid policy has some success mitigating this. No evidence is found for behaviour causing nodal price differences to change sign; as a result, the Hedge policy has a detrimental effect due to reduced competition. While the research presented in this thesis provides a necessary and useful first step towards a better understanding of strategic interaction in coupled markets, it is far from a definitive judgement on the issue. Further research is recommended to focus on one of the following areas: the generalisation of this work to include a more dynamic power market; an empirical analysis of existing combined FTR-electricity markets; or a comparison to the susceptibility of other congestion management schemes to strategic behaviour.