Luis Mundaca

Energy efficiency scenarios are developed in the national and international context to explore and evaluate different policy designs and visions of how energy will be generated, distributed and used in the future. However, these scenarios are often developed using conventional bottom-up modelling tools that, to only a limited extent, take into account decentralised decision-making frameworks, such as household investment decisions regarding energy-efficient technologies. The tools for modelling policy evaluation need to be improved to capture the factors determining the choice of technologies that affect household energy consumption and how these might be better influenced by means of energy efficiency policy instruments. In this paper, we present the first phase of a project analysing possible options to further improve microeconomic decision-making frameworks for evaluating energy efficiency policies and developing more realistic energy use forecasts for the household sector. The objective of the paper is to identify and explore a wide range of determinants beyond the narrow but traditional ‘rational model’ technology choice approach affecting and influencing households’ purchase/investment decisions regarding energy-efficient technologies. Furthermore, and within the economic/engineering paradigm that dominates energy modelling tools, we focus on the specific, but relevant issue of discounting to simulate and assess household preferences regarding energy-efficient technologies. Based on an extensive literature review, we present a summary of the body of evidence developed in the field. The results show that capital and operating costs prove to have an important influence on technology choice. However, the evidence clearly suggests that a broader set of determinants need to be considered and that different determinants will influence households’ technology choice in different markets under different circumstances and for different technologies. Even if pure economic parameters are examined, there is still a gap between what ex-post analyses reveal and the discount rates used in ex-ante modelling exercises. The results suggest that a larger representation of determinants in energy modelling tools is necessary to further enhance our understanding of household technology choice and thus the feasibility of such models in policy evaluation.