The essential role of scenarios in navigating energy market uncertainty

Why do we model scenarios?

The energy system is complex, and a future pathway cannot be predicted with absolute certainty by any one organization or person. Within this uncertainty, quantitative views of the future are usually constructed using computer software, optimization algorithms and some combination of the following approaches:

• Starting from today and following defined “rules of the game”, where does the system evolve to at a given point in future?
• If an end point target is known, say “reach net zero emissions by 2030”, then what needs to happen between now and then to achieve this target?

But the energy system is an intricate ecosystem of stakeholders, including generators, consumers and distributors, pursuing their own objectives within frameworks of markets, regulations and policies. Changes can happen across timeframes of hours, days, months, or years. Even the most advanced computer models cannot possibly explore all possibilities arising from this ecosystem, and hence scenarios act as a way of studying the future in a manageable way. Having a small set of different scenarios that are all modeled in a consistent way, using the same underpinning foundation of key drivers helps stress test decisions under diverse but plausible conditions. With energy markets continuing to evolve in the presence of geopolitical tensions and technology development against a background of needing to reduce carbon emissions, understanding scenarios has never been more important.

What makes a good scenario?

A good scenario should have three core characteristics:

1. Credibility: A scenario should reflect an outcome that is agreed to be within the realm of possibility - most importantly, it should not be impossible. It should also have an overall narrative that is credible, demonstrated by a scenario foundation that can be described in simple, understandable terms with outcomes that are explainable relative to the foundation and the modeling approach taken.
2. Distinctiveness: Scenarios should share a common framework and be based on similar drivers, but they should lead to differentiated outcomes that add value in promoting analysis and understanding. They should not be too narrowly defined in order to mutually reinforce a pre-existing viewpoint or desired outcome.
3. Consistency: Scenarios should be built on robust modeling processes that are internally consistent. This means taking into account relationships between factors (e.g., investors need to ensure profitability before they will build) and maintaining “real world” requirements (e.g., the need for back up plant) that often make the most mathematically elegant solutions redundant.

Why would a new scenario need to be created?

Good scenarios must maintain relevancy; hence they will be consistently updated. This constant evolution of input assumptions and modeling approaches can help to keep scenarios “fresh” and in line with market, regulation or policy movements, but sometimes a greater change is required. When the range of possible outcomes for a core scenario driver has changed over a long period of time, say due to ongoing price volatility or geopolitical uncertainty, then a new exploration is required. Adding a new scenario presents an additional viewpoint that can be used by stakeholders, challenging them to think more widely about what they “believe” and what the consensus across the energy sector “believes”.

Scenarios vs sensitivities – what level of “shocks to the system” do scenarios cover?

Although there is no single definition of a scenario, energy system scenarios generally focus on pathway developments over ~30-year horizons, so there is time for planned action and reaction within the modeled system. Simply put, if energy demand is expected to increase, then new generating capacity will need to be built or sourced. If the cost of a particular generation type rises, then this option will become less cost competitive and will, over time, be replaced by other technologies. Scenarios also account for the types of disruptions that system operators plan for in the real world. For example, “security of supply constraints” can be enforced to ensure there is sufficient “back-up” generating capacity to cover for a lack of generation due to maintenance or unplanned plant outages. In a world of increasing renewable generation, detailed weather analysis is also vital to ensure that there is no over-reliance on the sun shining and the wind blowing at assumed typical daily and seasonal levels.

Scenarios, however, are not good at accounting for unexpected “shocks to the system”, particularly at the time when the shock first happens. Shocks can refer to the impacts of extreme weather events, war and conflicts, financial and market crises, and disease pandemics. The key features of shocks are that they often have no direct historic precedents and are associated with high uncertainty, both in terms of what the impact might be and how long the impact might last. What this means practically, is that “optimal” solutions do not exist in the immediate energy system response, creating enormous volatility. Disrupted supplies may not have ready replacements available, so the reaction is most likely to be significant prices rises.

Shocks are therefore better studied through “sensitivity analysis”. This still examines the question “what happens if …”, but typically only changes one input of the underlying scenario at a time. This is a simpler approach compared to scenario construction, but the advantage is that it allows quick analysis to topically relevant questions, such as the impact on power prices if gas prices rise by +20%, for example. Any persistent impacts from a shock (e.g. changes to behavior or policies) should then be used to further refine existing scenarios to keep them relevant and credible.

Why AFRY?

Scenarios only become meaningful when they’re built on a consistent and transparent foundation. At AFRY, we bring together all major system drivers, from global commodity markets and technology trends to electricity demand shaped by EVs, heat pumps and data centers, within a single coherent modeling framework. This systemwide approach helps clients explore how different factors interact across future pathways, rather than analyzing each trend in isolation.

Different organizations engage with scenarios in different ways, so we offer a range of routes to insight. Many clients use our market reports to track evolving assumptions and outcomes across regions, while others work with us directly to interpret results, challenge assumptions or explore sensitivities in more depth. For those wanting to build or adapt their own internal modeling capability, our BID3 software can also be licensed and used within their own teams.

Whatever the approach, our focus is the same: helping clients understand the implications of uncertainty and make more informed strategic and investment decisions.