The short answer:
There is no perfect model. The 2050 pathways model is playable & transparent but simplistic. Markal, ESME, the DECC Energy Model are rich & powerful in different ways but complex and limited in others. DECC uses all four.
The long answer:
There is no perfect model.
The 2050 pathways model tries to be:
- Just complicated enough to give the gist of what could happen to the UK energy system over the next 40 years.
- Just simple enough that a person can explore the possibilities in a few hours.
- Just rich enough to give people the freedom to try out their own beliefs (which may mean, for instance, that they don't necessarily want what is cheapest)
- Transparent, so that the public can check the calculations and the assumptions - and modify both if they disagree.
- A forward looking tool - it does not solve for a specified end-point target, rather explores the implications of choices made
But this means the model has limitations:
- It doesn't follow a specific framework, whether a cost-optimizing framework or a forecasting model - it is a tool to explore various trade-offs within an energy system
- It doesn't find the cheapest answer - it has no automatic optimiser.
- It doesn't predict - it is driven by the assumptions and biases the user makes in their choices.
- It can't be used to test the effect of policies - it does not have policy feedback mechanism built into it, and has no view on how people will behave.
So, when considering what to do, DECC uses a combination of models and analysis.
DECC uses many models and pieces of analysis for specific sectors and policies, and three main models that simulate the whole energy system: the DECC Energy Model, Markal and, more recently, the ESME model developed by the ETI.
The DECC Energy Model
- The DECC Energy Model aims to be the best at providing forecasts of the effect of policies
- The Markal model aims to find the least cost pathway of the total UK energy system given an emission constraint, it assumes perfect foresight and perfect markets
- The ESME model tries to incorporate the implications of the uncertainty in the cost of technologies