Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Modeling market power in electricity markets is fraught as agents compete in prices but interact daily. In deciding what supply to offer, generators need to form judgements on the supplies chosen by rivals and hence the residual demand they face. Many markets are found to have prices above competitive levels, which could be explained by Nash-Cournot behaviour or marking-up above variable costs, but these strategies may not be robust against sophisticated deviants. This paper demonstrates that (1) the Nash choice of the optimal proportional mark-up on marginal costs yields lower prices and profits than Cournot behaviour but higher prices and profits than the optimum fixed mark-up; (2) such mark-up models are robust to single firm Nash deviations, but not against more sophisticated deviations in the deterministic case, nor under demand uncertainty. Proportional mark-up models emerge as the most robust and hence preferred modeling approach.