To improve the evaluation of electricity sector policy instruments before their implementation, ESS developed a model of two interconnected electricity markets in which investment decisions are made by individually modelled energy companies. A variety of policy instruments can be ‘turned on’ or ‘off’ in this model, for the entire market or for only one region in an interconnected system, in order to test their real-world effects.
Our agent-based approach to long-term modelling is fundamentally different from other models. The advantage of our approach is that it enables the simulation of realistic investment behaviour, which is characterised by a strong physical path dependence and limited knowledge of the future. We use this simulation model to analyse policy options for the electricity sector.
The agent-based approach makes it possible to simulate the impact of policy measures on investment behaviour, even if investors are confronted with multiple policies at the same time, such as CO2 policy, renewable energy policy and a capacity mechanism. We can also simulate the cross-border spill-over effects of policy instruments, which is currently a concern with respect to the implementation of capacity mechanisms.
The project intends to improve the quality of policy interventions. Our approach of simulating realistic policy measures under realistic circumstances is a departure from the traditional optimisation models that are frequently used for policy analysis, as we are not restricted to the same, limiting assumptions regarding, for example, optimal investment decisions. The benefit to society is insight into the possible consequences of policy instruments, intended and, importantly, unintended ones. The latter – the possible downsides – traditionally remain underexposed, as they are often caused by factors that are not modelled.
A key example of imperfect market behaviour is the tendency of electricity markets to develop an investment cycle. Conventional equilibrium models do not capture this. As a consequence, these models are blind to the risk that a period of overinvestment in generation capacity – as is the case in the Netherlands now – is followed by a period of underinvestment. Our model shows that this tendency towards an investment cycle will become stronger as the share of solar and wind energy increases. This means that the debate about whether to intervene by, for example, implementing a capacity remuneration mechanism, will probably keep returning. (France and the UK have implemented a capacity market; the Netherlands and Germany have decided against it for now.) Our model shows the possible effects of different types of capacity remuneration mechanisms, including possible cross-border effects.
Our model also shows how the investment cycle in the power market contributes to an investment cycle in the CO2 market. We have presented the research to the European Commission and the Dutch Ministry of Economic Affairs. The CO2 Market Stability Reserve was much improved along the lines that we suggested, but it is still too early to tell the extent of our impact on the decision-making process.
‘We intend to revolutionise the way in which energy policy instruments are evaluated by providing a realistic environment for testing the effects of energy transition policies.’
Who is involved?
|Department of Engineering Systems and Services|
Laurens de Vries, Émile Chappin, Jörn Richstein, Pradyumna Bhagwat, Kaveri Iychettira, Salman Khan
How is the project funded?
Through the Sustainable Energy Systems and Technologies Joint Degree Programme (Erasmus Mundus EU scholarship programme).