Modellingenergydemandinenergysystemmodels

The issue: a need to better understand the models that inform demand-side energy policy-making

Under the UK Climate Change Act 2008, the government is legally bound to reduce greenhouse gas (GHG) emissions by 80% by 2050 relative to 1990 levels.

Historically, the focus of energy policy in the UK has been on supply-side policies, such as decarbonisation of electricity generation through greater use of low carbon technologies like wind and solar. Increasingly, however, demand-side energy policies are being recognised as having important contributions to make to achieving emission reduction targets, through reducing energy demand or by making energy demand more flexible and compatible with variable renewable energy sources. Such demand-side policies can seek to promote a wide range of technologies and behaviours, for example improved building insulation, reduction in the use of energy intensive materials and increases in teleworking to reduce commuting.

To fully realise the potential of demand-side energy policies, it is important that they can be adequately represented in quantitative energy models, because such models play an important role in informing UK energy policy. However, we do not currently have a good understanding of how well the different energy models that inform UK government energy policy represent energy demand and demand-side energy policies.

Therefore we have undertaken a Rapid Evidence Assessment (a constrained form of systematic review) to examine the energy models that have informed energy policy documents published by the UK government between 2007 and 2017. The overarching question this review seeks to address is:

How suitable are the energy models used to inform UK government energy policy for exploring the full range of contributions that demand-side energy policies can make to climate change mitigation? 

Main findings: strengths and weaknesses in existing modelling

Our Rapid Evidence Assessment reveals that the core strength of current energy modelling is the detailed representation of technologies, with many models featuring information on hundreds of potential technological options for increasing energy efficiency. Although uncertainties exist around these technological options, these models allow us to gain a coherent and realistic understanding of how different combinations of technologies could satisfy our future energy service demands under different low-carbon scenarios.

However, the modelling landscape reveals two key limitations with regard to the representation of non-technological drivers of energy demand:

• Firstly, economic, social and behavioural drivers of energy demand, such as thermostat settings, are rarely explicitly included within the models. As a result, changes in such drivers risk being overlooked as potential levers of climate change mitigation. In addition, the economic, social and behavioural assumptions that are included into the models are often not presented in a transparent manner.
• Secondly, while many of the models provide a wealth of detail on different technological options, they are not well-equipped to model how behavioural, social and economic changes impact on the use of technology. This makes it difficult to investigate how technological change comes about and how it could be steered into the right direction using different policy instruments. In many models, the uptake of different technological options is specified exogenously by the user, or is determined within the model by cost optimisation, which does not reflect the many non-cost factors influencing technology diffusion.

Key recommendations for future modelling

• Develop consistent and transparent processes for estimating and presenting the fixed inputs to energy models, such as energy service demands and socio-economic drivers.
• Quantify the potential of demand-side policies and include them as explicit options in energy models to make such policies more visible.
• Improve the representation of economic, social and behavioural processes in energy models where feasible and useful. This endeavour can build on existing research in the academic literature, which is currently not well represented in policy development.
• Research further the potential implications of interactions between different drivers of energy demand and make sure they are consistent. An example of such an interaction would be the impact that a modal shift from cars to cycling would have on the growth and energy demand of the automotive industry.

Download the working paper here

Scoping note: Modelling energy demand in energy system models