a b s t r ac t Untapped energy savings potential in the residential sector might lead to substantial welfare losses. While several studies have focused on the role of behavioral biases in explaining the lack of adoption of energy-efficient durable goods, little is known about the role of limited energy-specific knowledge and financial literacy. In this paper, we propose an integrated concept of 'energy-related financial literacy', which combines both energy cost-specific knowledge and skills

more »

... ded to process this information. Using data from a large household survey in three European countries, we explore the determinants of different measures of literacy and, most importantly, we provide empirical evidence on the association between limited knowledge and skills to perform an intertemporal optimization and the adoption of energy-efficient light bulbs. Our findings support the promotion of energy-specific financial education programs and tools to increase the adoption of energy-efficient durable goods. of consumer goods more generally, it has been observed that individuals are often inattentive or myopic with respect to additional costs associated with the purchase of goods, such as sales taxes (Chetty et al., 2009) or add-on products needed for future operation (Gabaix and Laibson, 2006) , which has also been observed for operating cost of energy-using durables (Andor et al., 2020; Houde and Myers, 2019) . Although many investments in energy-efficient electric appliances, heating systems or energy-efficient home renovations ensure net savings over their lifetime (McKinsey & Company, 2009) , households often fail to make these investments (Allcott and Taubinsky, 2015; Houde and Myers, 2019) . In the literature, this is referred to as the energy efficiency gap (Jaffe and Stavins, 1994) . Various market and behavioral anomalies have been discussed in the literature as potential determinants of the energy efficiency gap (Broberg and Kazukauskas, 2015; Gillingham and Palmer, 2014) . For instance, refer to the presence of salience bias, biased beliefs, endogenous inattention and present bias. What is specific to optimal investments in energy-using durables is that they require not only awareness of future operation costs but also a combination of specific knowledge and skills: information about the energy consumption and the lifetime of appliances as well as of possible new, more efficient, appliances that could replace the old ones. Furthermore, consumers need to know the cost of electricity and make assumptions on how frequently they plan to use their appliances. Eventually, they need the skills to process all this knowledge in order to identify the possible savings from replacing their appliances by more energy-efficient ones. Blasch et al. (2017b) show that these processing costs are relatively high for a substantial share of individuals. The lack of energy-specific knowledge and skills might represent an important barrier to households' energy conservation behavior, together with other internalities studied in the literature (Cattaneo, 2019; Allcott, 2011b; Andor and Fels, 2018; Fiorillo and Sapio, 2019) . A relatively large body of literature has recently developed around the role of financial literacy and its importance of economic decision making in several contexts (Hastings et al., 2013; Lusardi and Mitchell, 2014) . Brent and Ward (2018) extend this literature on financial literacy to the energy domain and test in a stated preferences experiment whether an augmented measure of financial literacy -including the "Big Three" questions and questions on the lifetime cost and payback periods of alternative hot water systems -increases the willingness to pay for energy efficiency, and hence increases the chances that consumers reap the benefits of investing in efficient appliances. Other studies have considered the role of different measures of energy-related knowledge, usually referred to 'energy literacy', on households' electricity consumption and energy conservation behavior (Brounen et al., 2013; Kalmi et al., 2017; Blasch et al., 2017a Blasch et al., , 2019 Blasch et al., , 2017b . The US government defines energy literacy rather broadly as 'an understanding of the nature and role of energy in the universe and in our lives' as well as 'the ability to apply this understanding to answer questions and solve problems.' (U.S. Department of Energy, 2017, p.1) and explicitly acknowledges that an individual's energy literacy has consequences for economic outcomes. However, a common understanding of how to measure energy literacy has not developed yet. Part of the literature proposes a broad concept of energy literacy that focuses on energy-related knowledge, attitudes and behavior Powers, 2011, 2013; van den Broek, 2019), whereas other studies propose a concept of energy literacy based on lifetime cost calculations (Brounen et al., 2013; Kalmi et al., 2017; Blasch et al., 2017a Blasch et al., , 2019 Blasch et al., , 2017b Filippini et. al. 2020) . In the first part of this paper, we therefore summarize the various possible definitions of the term 'energy literacy' as well as its relation to the concept of 'financial literacy' (Lusardi and Mitchell, 2014) . Moreover, we propose an integrated concept that we call 'energy-related financial literacy'. This concept combines both (1) the energy cost-specific knowledge households need in order to take informed energy-related decisions and (2) the set of skills needed to process this information, which is comparable to the set of skills needed for intertemporal investment decisions like pension planning. We propose that the combination of these elements predicts rational energy-related investment decisions better than a pure financial literacy measure, which we verify based on survey data on actual technology adoption from three European countries. Using data from a large sample survey carried out in Italy, the Netherlands and Switzerland, we present evidence about the level of financial literacy (as measured with the "Big Three" questions as in Lusardi and Mitchell 2014) and our measure of energy-related financial literacy. We document a substantial lack-of energy cost-specific knowledge among the respondents in our sample. We further perform two econometric analyses: First, we analyse the determinants of both literacy measures using multivariate regression analysis, with particular emphasis on the role of gender. The results of the multivariate regression analysis show the existence of a substantial gender gap in energy-related financial literacy, consistently with previous findings for financial literacy (Almenberg and Dreber, 2015; Lusardi and Mitchell, 2014) . Finally, we explore the role of financial literacy and 'energy-related financial literacy' for the adoption of energy-efficient technologies by exploiting data on actual investment decisions in durable goods. We consider the decision of consumers with respect to the adoption of energy-efficient light bulbs, which are especially interesting due to the large relative efficiency gains that consumers can achieve. The empirical analysis shows that energy-related financial literacy is positively associated with the lighting efficiency at home. Consumers with high energy-related financial literacy are associated with a 5 percent higher share of LED light bulbs at home. We do not find an influence of financial literacy on the adoption of lighting efficiency, which suggests that financial literacy alone is not sufficient to ensure optimal energy-related investment choices. This paper contributes to the literature in several ways. We discuss various existing concepts and definitions of 'energy literacy' and propose an integrated concept of 'energy-related financial literacy' that combines the set of knowledge and skills needed by consumers to take optimal decisions of investment in energy-consuming durable goods. We provide empirical evidence about the determinants of 'energy-related financial literacy', documenting an important gender gap and complementing previous findings for financial literacy. Finally, our study is the first to provide empirical evidence about the consequences of limited energy-specific intertemporal optimization skills for the actual adoption of energy-using durable goods. While our proposed measure of energy-related financial literacy captures similar dimensions as the augmented financial literacy measure used by Brent and Ward (2018), our study adds to the literature by investigating the role of literacy using revealed preference data on investment choices, rather stated preferences data. Further, compared to Blasch et al. (2017a) who consider the role of energy literacy and basic financial literacy separately on households' electricity consumption, we explore the role of a more comprehensive and integrated measure of literacy specifically on the adoption of efficient energy-consuming durables. Their results also suggest that financial literacy makes choices more consistent with a model of consumer behavior that assumes standard preferences. In a similar stated preferences setting, Andor et al. (2019) find that individuals scoring higher in a standard cognitive reflection test display a higher willingness to pay for energy efficiency. These findings suggest that, besides energy literacy, also financial literacy and high cognitive abilities positively influence rational energy-related choices. The overall financial implications of a low level of energy and financial literacy can only be estimated. Allcott and Taubinsky (2015) suggest that, in the year 2010 alone, not tapping the potential energy savings from replacing all incandescent by compact fluorescent light bulbs cost US consumers a total of 16 billion USD. As the total expenditure on energy-using durables is relatively high, Allcott (2016) assumes that consumers' inability to tap energy-efficiency potentials can sum up quickly to large welfare losses.