Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/238216 
Year of Publication: 
2021
Series/Report no.: 
Economics Working Paper Series No. 21/359
Publisher: 
ETH Zurich, CER-ETH - Center of Economic Research, Zurich
Abstract: 
For more than forty years analysts have pointed out that society might be too slow in adopting energy efficiency technologies, a phenomenon known as the Energy Efficiency Gap. There are persistent market barriers that impede these efforts. Eliciting these barriers and their heterogeneity is key for policy design. In this paper, we use narratives, a novel approach based on unstructured text answers in surveys, to elicit the barriers and determinants of energy efficiency investments. Using recent advances in Natural Language Processing (NLP), we turn narratives into quantifiable metrics to rank households' barriers and determinants. We find that financial motives are not the primary barriers or determinants of energy efficiency investments. Instead, we find that such investments are highly opportunistic and co-benefits, such as ecological concerns and comfort, also play an important role. Although there is substantial heterogeneity across the population in the type of barriers and determinants, demographics and building characteristics poorly predict heterogeneity patterns. This has important implications for the targeting of policies. Narratives could be a novel and effective way to implement policy targeting.
Subjects: 
energy efficiency gap
natural language processing
policy targeting
open-ended questions
JEL: 
Q41
Q50
L15
D12
D83
D91
Persistent Identifier of the first edition: 
Document Type: 
Working Paper

Files in This Item:
File
Size
945.96 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.