Can Enhancing Capabilities Promote Energy Justice? An Agent-Based Model Approach
The Capabilities Approach has been recently invoked in efforts to study questions relating to energy and justice, enabling a better understanding of the relationship between energy and well-being or deprivation. This framework allows for a flexible analysis of human needs in their local context and suggests a variety of possible interventions, beyond merely increasing energy consumption. At the same time, research in other areas such as decarbonization as well as environmental justice has derived important insights from agent-based models (ABMs). ABMs are especially useful in studying complex phenomena as they allow for a bottom-up interactions of heterogeneous agents with simple behavior rules, and give rise to emergent properties at the system level which are not reducible to any individual agent’s action or intent. This approach is thus very suitable for studying questions relating to energy justice, and is employed in this paper by creating an agent-based model of energy and capabilities. We study the differential needs of people for an energy service (heating) and their effect on a secondary capability (thermal comfort) and a basic capability (health). Furthermore, we look at several conversion factors – both personal and environmental – which mediate the transformation of commodities (e.g. energy services) into capabilities. We then utilize the model over numerous iterations to study questions such as whether equalizing agents’ secondary and basic capabilities can contribute more to energy justice than merely raising the floor of the most vulnerable in terms of energy consumption (a version of Sen’s question “inequality of what?”), and also look at whether such an approach is linear or has a threshold effect. Finally we analyze the results of the simulations for alternative ways of enhancing both human capabilities and ecological functionings, thus promoting sustainable energy justice. The attached files incolde the NetLogo code, as well as output data from the various scenarios described in the paper.