Global electricity technology substitution model with induced technological change

Future energy planning which aims to avoid excessive radiative forcing due to anthropogenic greenhouse gas (GHG) emissions leading to a global warming of more than 2 degrees C is likely to require drastic reductions in greenhouse gas emissions, possibly an almost complete decarbonisation of the current global energy sector. Such a transformation is expected to involve drastic costs, and large uncertainties surround the concept of decarbonisation and as to whether it is feasible economically. The transformation of the energy sector is likely, therefore, to have major consequences on the global economy, and it is difficult to model the energy sector without including its interactions with global economic activity. E3MG is a disaggregated global macroeconometric model which features an electricity technology submodel, involving a powerful combination of top-down and bottom-up approaches to power systems modeling.

However, this submodel currently lacks a treatment of natural resources, and does not reproduce adequately some of the important dynamics underlying changes in technology and energy infrastructure.

We propose in this work a novel approach to electricity technology substitution modeling as a development of the electricity submodel of E3MG. As opposed to traditional energy models based on cost optimisation procedures, it focuses instead on the dynamics of technology substitution in connection with induced technological change. Technology costs are influenced by learning-by-doing effects, which lead to strong path dependence. The model is designed to work with several world regions and thus with local energy landscapes. These regions are defined by the availability and costs of natural resources.

Preliminary calculations using a single world region are given in order to explore the properties of the model given very simple sets of assumptions.

The results highlight how technological change dynamics emerge from the set of equations at the root of this model.

Mercure, J. - F.

Ideas & insights