How to tackle energy efficiency in transport?
13 September 2023
The ESCO Market in Italy and Germany; Overview and Areas of Improvement
13 September 2023
By Giulio Nicoletti
Energy efficiency is a crucial element of the European Green Deal and the European decarbonisation objectives. (Delivering the European Green Deal, 2021) Yet, despite recognition of energy efficiency policies both at the EU and Member States’ level, the adoption and investment in energy efficiency measures remain low, particularly in the building, industry, and transportation sectors (European Commission, Joint Research Centre, 2021a). This inadequate uptake of investments, despite the clear economic benefits, (IEA, 2019) has sparked a discussion in the literature on potential barriers to energy efficiency and the existence of an investment gap in this sector. Both market failures or behavioural anomalies and failures can explain this phenomenon. Market failures arise when the market does not efficiently allocate resources due to various factors, such as information asymmetries, externalities, and regulatory problems. On the other hand, behavioural failures occur when decision-making and behaviour of individuals deviate from rational economic models due to cognitive biases, societal norms, emotions, and psychological factors. These behavioural factors may also explain suboptimal outcomes in energy efficiency investments, despite clear economic benefits.
This paper aims to review evidence of non-monetary behavioural barriers that restrict investments in energy efficiency, namely individual time preferences and present bias over effort. It will examine findings by Newell and Siikamaki (2015), Lades, Clinch and Kelly (2021), and Fowlie, Greenstone and Wolfram (2015) to evaluate whether the empirical results on biases are robust and applicable at the EU level. Secondly, this paper will examine preliminary evidence by Trianni and Cagno (2012) to examine whether these biases may also be applicable to micro, small and medium-sized enterprises.
Individual time preferences
One of the biases that has been tested empirically as a significant barrier to investment uptake in energy efficiency is time preferences. Newell and Siikamaki explored evidence of time preferences by drawing evidence from a pool of 1,217 randomly selected individuals in the United States. Each participant in the study took part in computerized surveys involving several decisions in which they had to choose between preferred products from three different water heaters, considered as an important energy efficiency technology for households. The experiment entailed 12 randomized label treatments that altered the energy efficiency information available to them. To determine whether individuals overvalue or undervalue energy savings, researchers estimated people’s willingness to pay for energy efficiency by using data from the survey and a statistical model called “multinomial logit.” This model helps to understand how people make decisions when faced with several options. The model analysed the reduction in the present value of operating costs, which is the current value of the future energy costs saved by investing in energy efficiency. As a preliminary result, the study finds that the label style has a strong influence on the valuation of energy savings, and that the lack of such information leads to the undervaluation of energy efficiency. A second experiment was carried out to estimate individual time preferences. Respondents had to choose between a $1,000 payment in one month or a higher payment in 12 months. Each participant who accepted the near payment was exposed to subsequent new questions with a larger 12-month payment, until they accepted. In this way, individual time preferences were examined as determinants of households’ willingness to pay for annual energy savings, combining data from the appliance choice experiments earlier described. The results are robust and statistically significant, showing that individuals with higher discount rates have a lower willingness to pay for energy-efficient solutions. This provides evidence that some individuals prefer immediate benefits over delayed benefits, even when the delayed benefits are larger or more valuable. This study also evaluated the average payback preferences of individuals, finding that for the water heater, respondents stated a mean and a median payback threshold of 3.5 years, with a standard deviation of 1.9 years for the investment to remain attractive. Finally, on the control variables, education is associated with a drop in the discount rate, meaning that more educated individuals are less prone to display time-preferences.
Present bias over money and present bias over effort
The study by Newell and Siikamaki, found evidence of time-preference biases associated with energy efficiency investments, but did not go as far as showing a breakdown analysis on why people procrastinate investments that are economically advantageous. Lades et al. (2021) attempted to address this gap by trying to understand whether this bias is due to monetary-biases -namely, people have biases over spending money today over saving tomorrow- or due to information/effort burden biases -namely, people have biases over informing themselves and putting effort on energy efficiency today over saving tomorrow. The authors looked for causal evidence by using data from the Irish housing stock for a specific energy-saving technology, Air Source Heat Pumps (ASHPs). This technology was chosen because it is recognized as particularly important for the energy transition and because it offers clear benefits of different natures to consumers: it improves comfort, air quality, health, and significant energy savings.
In particular, the authors used a dataset that collected data from 2009 to 2021 and analysed a subset of 376,417 households for which the investment was likely to be highly beneficial. The authors calibrated a model that allowed for the prediction of the share of households investing in ASHPs, using different assumptions about householders’ present bias and the size of administrative burdens. The model assumed individuals to make two sequential decisions: first, a household decides whether to make an energy efficiency investment, thus expressing an intention to invest, and then they need to decide when to make the necessary arrangements to implement it, thus weighing the decision against the costs of putting an effort into it and against administrative burdens, in terms of paperwork, phone calls, or others. In this setting, the authors approximated the savings potentials from ASHPs, assuming an investment cost ranging from €8,700 to €9,800, a maximum heat required between 8kW and 16kW, and a lifetime of appliances of 20 years. The simulation then assumed an administrative cost up to €50, calculated by considering an average hourly salary in Ireland and the time spent in the process, assumed to be around 2 hours. The simulation was run with three different present-bias parameters in line with the literature (between 0.7 and 0.9).
The results of the simulation predicted that in the absence of administrative burdens and with no government funding, 69% of households would benefit from the investment. Moreover, in the same case of no effort burden and a government grant of €5500, the uptake increased to 97%. However, in the case of an administrative cost of €50, the model predicted only a share of 7% of households to invest in ASHPs, a number that reached 10% in the presence of the same government grants of €5,500. This shows that monetary incentives are particularly effective only when efforts are lowered. The authors argue that the model’s logic is applicable to several technologies with saving potentials.
Reducing investment burdens: an experiment
A different empirical study, though, shows that present bias over effort may not be easy to eliminate. Fowlie, Greenstone, and Wolfram run a large-scale randomized control trial and data from the US Federal Weatherization Assistance Program (WAP), a program of energy efficiency dedicated to low-income individuals active since 1976, for which 7 million people received assistance. The program has been shown as effective in significantly reducing energy expenditures among participants (Fowlie et al., 2015). Participants received a free energy audit and a retrofit that included typically insulation, window replacements, infiltration reductions, furnace replacements, for an average value of $5000. Households did not incur in direct monetary costs, but the application process was particularly burdensome (the government wanted to prevent frauds), as participants must submit significant paperwork that includes earning documentations, utility bills, social security numbers and others. The experiment run by the authors, in particular, concerned a sample of 30,000 households in Michigan. A treatment group, formed by one-quarter of the sample randomly selected was educated about the benefits of the program and received personal assistance for completing the application. A control group was free to apply to the WAP but was not assisted or contacted by the supporting team. The authors described the effort to reduce barriers to participation as “massive”, as 7,000 in-person house visits were made, 23,500 targeted calls, 15,000 door hangers and mailed post cards. In the enrolment phase, 9,000 phone calls and 2,700 call visits helped the individuals complete and deliver the necessary paperwork. The encouragement effort costed approximately $450,000. After the treatment, only 15% of the treatment group made the application, and fewer than 6% received the WAP. In the control group, 2% applied and 1% received the programme. One possible explanation of such a low uptake could stem from the fact that receiving the call and accepting to speak and get help from a supporting team still requires an effort, that has not been factored in in the analysis. Yet, the findings are robust and relevant, showing that effort bias is particularly difficult to overcome. Also, the breakdown analysis offers some interesting insights. First, the information campaign is particularly effective for individuals with higher income and for larger households, that are thus more likely to have children at home and more likely to have an elderly resident. Reduced information costs are also found more relevant for households with disabled individuals.
Behavioural biases: expanding the scope to MSMEs?
After finding insights on possible biases regarding individual decision-making decisions of investment in energy efficiency, it is relevant to question whether these biases may also apply to micro, small and medium enterprises (MSMEs). Most of the studies in fact explored the barriers behind underinvestments in energy efficiency at individual levels, without looking for possible applications to MSMEs. Trianni and Cagno run an experimental study on energy efficiency investments barriers of MSMEs in Italy, Lombardy region. Through a sample of 128 MSMEs, they run a survey with question with a Lakert scale to understand what problems were more relevant to energy efficiency investment decisions. The results point at slightly different directions compared to findings of studies at the individual level. In fact, 42 out of 128 MSMEs considered the lack of capital as a very critical barrier to energy efficiency investment. This suggests that MSMEs might deviate less from the assumptions of neoclassical economics compared to individual, with a classic monetary barrier explaining underinvestment. On the other hand, this does not mean that behavioural biases do not play a role. In fact, 18 MSMEs considered lack of time or other priorities (in comparing the energy efficiency efforts with respect to production efforts) as an important barrier, and more than 70% of the respondents considered poor information as an important barrier. While these findings do not show concrete evidence of biases such as present bias over effort or bounded rationality, they indicate that such biases are worth being considered in future research on the topic, being in line with such responses. Another interesting finding is given by the breakdown analysis of the results. Lack of time and poor information over energy efficiency is in fact found particularly relevant for the smaller enterprises in the sample. This may again support the idea that the smaller the enterprise, the more likely energy efficiency decision fall within a single individual, the higher the probability that time preference and present biases apply. This also suggests that in future studies it might be better to unbundle micro-small and medium-large enterprises when exploring behavioural biases in energy efficiency.
Critical analysis – Information gap
These analyses offer valuable insights and specific evidence of the role of present biases in individual energy efficiency investments. Yet, it is clear that there is still an important information gap regarding energy efficiency uptake at the EU-level. One limitation of these studies is that they were carried-out in specific settings that do not necessarily apply at the EU-level. Another limitation is that these studies are not aimed at comprehensively testing multiple biases, and do not explore the roles of social norm or other biases. Such limitations are relevant and highlight the need for future research. Yet, the studies give important preliminary findings to policymakers. First, the results are robust and significant, and while the studies by Newell et al. and Fowlie et al. were set in the United States, the results are in line with well-established behavioural economics principles. At minimum, these findings indicate that behavioural factors are indeed worth being considered by policymakers. Designing policies without considering behavioural factors will probably lead to suboptimal results. In particular, monetary incentives without proper information and without a system that reduces individual’s needed efforts to uptake energy efficiency solutions, are likely to be ineffective.
Second, there is a considerable information gap regarding behavioural biases in MSMEs energy efficiency investment decisions. The study from Trianni e Cagno, while not specifically testing behavioural biases, showed clearly that informational problems and the time and effort needed to invest in energy efficiency could be a barrier even for MSMEs, especially for micro and small enterprises where few individuals take the investment decision. Thus, their findings reinforce the idea future research in behavioural studies is needed to have more effective, evidence-based and efficient policies.
BIBLIOGRAPHY
European Commission. (2021) Delivering the European Green Deal. https://commission.europa.eu/strategy-and-policy/priorities-2019-2024/european-green-deal/delivering-european-green-deal_en
Fowlie, M., Greenstone, M., & Wolfram, C. (2015). Do Energy Efficiency Investments Deliver? Evidence from the Weatherization Assistance Program. Unpublished.
IEA (2019), Multiple Benefits of Energy Efficiency, IEA, Paris https://www.iea.org/reports/multiple-benefits-of-energy-efficiency, License: CC BY 4.0
Lades, L. K., Peter Clinch, J., & Kelly, J. A. (2021). Maybe tomorrow: How burdens and biases impede energy-efficiency investments. Energy Research & Social Science, 78, 102154. https://doi.org/10.1016/j.erss.2021.102154
Newell, R. G., & Siikamäki, J. (2015). Individual Time Preferences and Energy Efficiency. American Economic Review, 105(5), 196–200. https://doi.org/10.1257/aer.p20151010
Trianni, A., & Cagno, E. (2012). Dealing with barriers to energy efficiency and SMEs: Some empirical evidences. Energy, 37(1), 494–504. https://doi.org/10.1016/j.energy.2011.11.005