Last fall, Multiconsult Norge AS and THEMA Consulting Group surveyed uncertainty among power market analyses on behalf of Norway’s Energy Commission. The work shows broad consensus within these groups as to both the drivers and uncertainties involved.
The study shows that there is significant expertise in modelling the power sector in Norway, based on decades of experience with detailed modelling of hydropower and the power market.
Most of the analytical groups in the study use so-called “bottom-up technology-rich optimisation models”. These models optimise system investment and operation to minimize total costs and meet energy services demand. They are highly detailed, modelling all production technologies, in some cases down to the level of individual power plants.
However, despite their detailed representation of the power sector, in some situations, they can be too narrow in their scope, notably due to their limited consideration of substitution opportunities in both production and consumption. In general, the modelling and analysis is significantly influenced by the analysts’ own assessments.
There is a broad consensus among analysts as to the most important drivers, as well as those drivers subject to the greatest uncertainty. The drivers that are important and/or uncertain differ somewhat between the short and long term, but again there is broad consensus as to these differences.
Through the interviews, it was made clear that one cannot assess a driver in isolation, as they are closely interlinked. Some underlying drivers influence multiple others, and many drivers are mutually interdependent. We also observe that the interview discussions are influenced by the current situation, notably the war in Ukraine and the energy crisis in Europe, especially concerning the short term. The relevance of different drivers also changes with the expected development of the power system.
An obvious question is whether the extent of the consensus and the similarity in the models used is itself a problem. It may result in an overly narrow range of projected outcomes and suggest that analysts fail to adequately “think outside the box”. In particular, outcomes may not be “stress tested” to account for less likely extreme scenarios. It should be noted that sensitivity and scenario analyses are carried out, which capture a larger range of outcomes.
Also, if the analysts are transparent about the assumptions they have made, then users of the analysis can challenge these and make their own judgements as to the likely outcome space.
We also see that the focus on drivers can shift over time to reflect new information and recent developments. This suggests that projections are dynamic and that the actors involved can absorb new information and change their view on the most relevant drivers over time.
The mapping is based on interviews with different existing Norweigan analytical groups, complemented by a literature review. We interviewed 13 actors and reviewed 9 reports.
The report was prepared for the Norwegian Ministry of Oil and Energy in collaboration with Multiconsult.
