Energy Efficiency Policies in the Residential Sector

Resumo

The present thesis studies the diffusion of energy efficiency (EE from now on) in the residential sector using an economic perspective. The research is concentrated on the analysis of the Energy Efficiency Paradox and on the evaluation of EE policies. The thesis consists of five independent chapters (including this chapter), some of which have been already published in specialized journals. Although increasing EE in the residential sector has been for long time one of the targets of many governments, several reasons have led governments to intensify their efforts in this area. The negative effects generated by climate change, urban pollution and fuel poverty, together with the growing political instability in energy-producer countries, are some of the reasons that are behind this renewed interest both in developed and developing countries. Moreover, engineering studies have identified the residential sector as the one with the largest potential for cost-effective energy-savings. Despite the fact that mathematical models have identified “win-win” situations for households if they adopt EE measures, and the large number of public initiatives to promote EE applied by public institutions, the diffusion rate of EE continues to be lower than expected. Thus, this field of research has gained much attention, not only from policymakers but also from the society and academics. Despite the fact that engineering models have identified “win-win” situations for households if they adopt EE measures, and the large number of public initiatives to promote EE applied by public institutions, the diffusion rate of EE continues to be lower than expected. Thus, this field of research has gained much attention, not only from policymakers but also from academics and other stakeholders. Up to now, most of the interventions were oriented to internalize the negative externalities of energy through the use of price instruments (e.g. through Pigouvian taxes) or codes and standards that ensure a minimum level of energy performance. However, the empirical evidence has shown important limitations associated with such policies. They could be partly given by the general assumptions about perfect information and rationality that these models use. Indeed, much attention has been recently given to the ideas of behavioral economics, which has identified several situations when consumers systematically deviate from the rationality assumptions. The inclusion of behavioral economics approaches in the field of EE has opened the door to new interpretations of the Energy Efficiency Paradox but also to non-price interventions as an important complement to promote EE. Additionally, the development of experimental techniques has allowed researchers to evaluate the effectiveness of non-conventional public interventions. Therefore, the present thesis has the double objective of understanding household decisions regarding EE and of evaluating novel instruments to promote EE. On the one hand, with this research I tried to contribute by providing new academic evidence on the factors driving consumers’ choices. This first step is essential to explain the divergence between the level of investment in EE predicted by engineering models and the real level of EE (previously mentioned as the Energy Efficiency Paradox). Besides that, given that EE in the transport sector presents several parallels with dwellings, this thesis also analyses consumer’s decisions related with vehicles’ energy efficiency. By doing so, I verify whether the results can be applied to other fields of consumer decision related with EE. On the other hand, my research tries to evaluate the results of informational instruments, in particular energy performance certificates or labeling systems, now a widely-used tool to promote EE that has experienced a rapid diffusion in the last few decades. In this sense, Chapter 2 can be seen as an introduction to this thesis. The objective of this chapter was to acquire the necessary knowledge that provided me with a complete overview of the field of EE. This task consisted of the revision of a large number of studies and reports from different research lines: energy demand, fuel poverty, behavioral economics, electricity markets, environmental economics, etc. Furthermore, this chapter works as an introduction and serves as a guideline for the rest of the doctoral thesis. Firstly, it contextualizes the research on EE in the residential sector from an economic point of view. It updates the current debate about the Energy Efficiency Paradox with special attention to the growing role that informational and behavioral failures play in explaining this gap. Secondly, the core of the chapter is a review of the empirical evidence on the effectiveness of the informational instruments that have been rapidly adopted by many governments during the last decade. In particular, the chapter focuses on energy certificates, feedback programs, and energy audits. As far as I know, this is the first study that gathers together the results from economic research in all the existing informational instruments. Results show that energy certificates and feedback programs can be effective, but only if they are carefully designed: yet energy audits seem to have little effect on efficiency. In addition, the chapter points out the large potential for new instruments as well as for combinations of existing ones. Finally, it identifies the experimental techniques implemented in the last few years, highlighting the robust approach that use large and random samples and have a large potential in for future policy evaluation in this field. Chapter 3 empirically evaluates the determinants that drive household decisions regarding EE adoption in Spain. The objective of this chapter is to get insights into EE in Spanish households. Several studies have previously evaluated the factors that determine households’ choices. However, Spain presents some specific characteristics that demand for ad hoc analysis: an aged population, a remarkable increase of purchasing power, ownership status or the size and composition of the stock of buildings, among others. Using information from a large and representative official survey, in this chapter I focused on the effects of socio-economic characteristics on the probability to adopt EE measures or energy-saving daily habits. By doing so I was able to identify the barriers that prevent consumers from taking EE measures, and the characteristics that make them favorable to invest. Since EE measures can show different up-front costs, I used information on well-differentiated measures: from energy-saving habits, such as lowering indoor heating temperatures, to low cost measures, such as low consumption bulbs, and higher cost measures, such as energy efficient appliances or double glazing. When dealing with the Energy Efficiency Paradox, there is a general concern with the heterogeneity present among households. In this chapter, I focused on the effects of household environmental attitudes and behavior as a potential explanation for differences in the rate of EE adoption. The chapter shows that households with eco-friendly behaviors are more likely to investment in well-differentiated EE measures as well as to steer daily habits towards energy savings. However, no effects were found for households with environmental attitudes based on stated willingness to pay to protect the environment. In addition to this, households belonging to higher income groups and education levels are more likely to invest in EE but not to adopt energy-saving habits";" while households with older members are less likely to invest in EE and show fewer eco-friendly habits. In sum, this chapter provides the first empirical results for Spain on the factors driving household choices regarding EE. Moreover, it contributes to the existing literature by giving insights on pro-environmental attitudes and behaviors. One of the justifications for the use of command-and-control policies is the hypothesis that consumers do not value EE. Hence, the objective of Chapters 4 and 5 was to study consumers’ willingness to pay for EE and their valuation for EE labels or certificates in dwellings and light duty vehicles, respectively. This is particularly relevant because EU legislation has introduced the mandate to apply labeling systems to differentiate products based on their level of EE. The objective of those regulations is to help consumers to take efficiency decisions in a market with imperfect information. In Chapter 4 I estimate the implicit price of EE vehicles, measured as those with label A or B. I have used two samples, one with all vehicles on sale in Spain containing official information on attributes and commercial prices of vehicles, and another where a subsample of these vehicles was selected and matched with the price of retailers. For this subsample, the method of “mystery shopping” was carried out by a specialized company. This approach consists in asking directly the retailer about the price of certain vehicles as if this process was part of a real acquisition. A main objective of the chapter was to test the hypothesis of changes in willingness to pay for EE using official prices and retailer prices. Additionally, I used the European labelling system for light vehicles, which classifies automobiles according to their relative fuel consumption levels, as a novel and alternative indicator for EE. The results of this chapter indicate that vehicles labelled A and B are sold at prices 3 to 5.9% higher than those with similar characteristics but lower energy-efficiency labels, using official commercial prices and retailer prices, respectively. By comparing this results with the present value of the energy savings, it can be seen that the use of retailer prices overestimates the value of EE. Although, there is sizeable research on this field, as far as I know this is the first evidence for the Spanish car market. Furthermore, our results fall in the range of magnitudes previously found. Finally, in Chapter 5 I estimate the implicit price of Portuguese dwellings rated as A, B, or C, keeping constant the rest dwelling attributes. Given the lack of official and complete databases for the stock of buildings, I have downloaded the information on dwelling sales from the web page of one of the largest real estate companies in Portugal, and constructed a unique database with complete information of dwelling attributes. This database includes general dwelling attributes but also information on the level of EE of the dwelling and its energy performance certificate. Moreover, since location is one of the most important factors when purchasing a dwelling, I have used spatial economic techniques to control for the exact location of each housing unit. The results show that Portuguese consumers positively value high certified dwellings. Despite the fact that the database used for this study includes information on a larger number of dwelling attributes than in previous studies, reported values are higher. This suggests that the valuation for energy performance certificates is higher in the Portuguese residential sector than that in other European countries. Yet, it can also be the case that our model cannot control for unobservable factors related with energy performance certificates, and this overestimates the results. Indeed, as other researchers have pointed out, there could be effects such as reputation or status that models cannot control. Thus, the chapter highlights the need for further research in this area even though there is an unfortunate lack of official information.