Behavioral Anomalies and Energy-related Individual Choices: The Role of Status-quo Bias.

• Author: Blasch, Julia
• Position: Report

1. INTRODUCTION

   The literature on the energy efficiency gap (Jaffe and Stavins, 1994; Greene, 2011; Allcott and Greenstone, 2012; Gillingham and Palmer, 2014), i.e. the observation that individuals forgo opportunities to make economically viable investments in energy-efficiency, provides explanations for such under-investment in energy efficiency. Various market and behavioral failures as well as behavioral anomalies that may impede energy efficiency investments in households have been identified. (1) One of these explanations is that individuals are biased towards the status quo (Schleich et al., 2016; Frederiks et al., 2015; Schubert and Stadelmann, 2015; Broberg and Kazukauskas, 2015). This means they have the tendency to either maintain their current situation (i.e. keeping their current stock of appliances or current level of home insulation) or to replace their current equipment by products that they have purchased previously, without considering more energy-efficient options. This phenomenon is more generally known as the status-quo bias and is defined as an individual's tendency to "[do] nothing or [maintain] one's current or previous decision" (Samuelson and Zeckhauser, 1988, p.7).

   According to Samuelson and Zeckhauser (1988), the potential underlying causes for the status-quo bias may include transition costs, uncertainty, cognitive misperceptions like loss aversion, and psychological commitment due to perceived sunk costs or other resource investments. Frederiks et al. (2015) suggest that psychological commitment may cause an over-consumption of energy services: once an electric appliance has been purchased, individuals with a strong sunk-cost fallacy may perceive it as beneficial to use the appliance as much as possible, in order to mentally amortize the initial investment. (2) The status-quo bias could hence affect residential energy consumption in at least two ways: (1) by making consumers keep their energy-using durables as long as possible, until wearout forces them to replace their equipment, and (2) by making consumers overuse appliances in an attempt to mentally amortize the investment cost. Of course, the first behavior could also be driven by other behavioral failures, such as bounded rationality (Simon, 1959; Andor et al., 2020; Houde, 2018) and a related low level of energy and financial literacy (Blasch et al., 2018b; Brent and Ward, 2018; Blasch et al., 2018a) or high subjective discount rates (Hausman, 1979; Train, 1985; Coller and Williams, 1999; Harrison et al., 2002; Epper et al., 2011; Bruderer Enzler et al., 2014; Min et al., 2014). Disentangling the different causes for under-investment in energy-efficient appliances is important as they have different policy implications.

   While the relation between status-quo bias and energy efficiency has thus far mainly been discussed as a theoretical concept (Schleich et al., 2016; Frederiks et al., 2015; Schubert and Stadelmann, 2015; Broberg and Kazukauskas, 2015), empirical evidence for the link between the status-quo bias and residential electricity consumption is lacking. Some of the existing studies investigate the influence of related concepts, such as loss aversion, risk aversion and time discounting, on the choice of energy-efficient appliances and energy-efficient renovations (Farsi, 2010; Qiu et al., 2014; Heutel, 2017; Schleich et al., 2019). Yet, apart from the recent work of Martin-Bonnel de Longchamp et al. (2018) on the role of the status-quo bias for the energy performance gap in low-energy buildings, we are not aware of a study that explored the role of the status-quo bias for energy efficiency investment decisions and consumption of energy services empirically. With this study we want to contribute to filling this research gap.

   Our paper aims to capture the relationship between the status-quo bias of the household head and both the age of the appliances stock as well as the (over-)consumption of energy services, while controlling for energy-related financial literacy as defined in Blasch et al. (2018a) (3) and a proxy for time preferences. (4) The empirical analysis is based on a large household survey conducted in three European countries: Switzerland, the Netherlands and Italy. As part of the survey, house-holds were asked about the age of their current appliance stock as well as the frequency of using the major home appliances. We could also link information collected with the survey with electricity consumption data provided directly by the households' utilities. Furthermore, the survey questionnaire included a set of six survey items that capture preferences for the status quo and loss aversion in various contexts, which is based on the Loss Aversion Questionnaire developed by De Baets and Buelens (2012). From these six items we created an index to measure the strength of the status-quo bias of survey respondents. The measures of status-quo bias we use in the analysis show similar associations with socio-demographic characteristics as incentivized measures of loss aversion (Gachter et al., 2010; Tanaka et al., 2010; Hjorth and Fosgerau, 2011).

   In the econometric analysis, we regress indicators of energy-related choices on our measure of status-quo bias, controlling for a rich set of individual and household characteristics. We find that our measure of status-quo bias is a significant predictor of both the age of home appliances as well as the level of consumption of energy services of a household. When an individual is biased towards the status-quo, the probability that the individual's household owns at least one electric appliance that is more than 10 years old increases by around 4 percentage points. Also the consumption of energy services of a household increases when the household head is status-quo biased: our indicator of bias towards the status quo is associated with an about 5.4% higher consumption of energy services. This is also reflected in the total electricity consumption of the respective households, which is found to be 6% higher than the consumption of households in which the household head is not status-quo biased. The results are robust to using alternative measures of status-quo bias at the individual level, as well as to controlling for the respondents' degree of risk aversion.

   The paper is structured as follows: Section 2 provides an overview of the literature on status-quo bias, loss aversion and their relations to investments in energy-efficient appliances as well as consumption of energy services. At the end of the section, a conceptual framework for analysis as well as testable hypotheses are derived. Section 3 explains the data set and our measure of status-quo bias. The results of our analysis are presented in Section 4 and discussed in Section 5. Section 6 concludes.

2. THEORETICAL BACKGROUND

   2.1 The status-quo bias in relation to loss aversion, the endowment effect and the sunk-cost effect

   The status-quo bias was first conceptualized by Samuelson and Zeckhauser (1988) as "doing nothing or maintaining one's current or previous decision" [p.7]. They discuss four potential causes: (1) Transition costs, which make the deviation from the status quo costly in itself, (2) Uncertainty in the decisions situation, which requires costly effort to investigate alternatives and their benefits (search and decision-making cost), (3) Cognitive misperceptions like loss aversion (endowment effect), anchoring or bounded rationality, and (4) Psychological commitment due to perceived sunk costs or other resource investments or due to regret avoidance. Samuelson and Zeckhauser (1988) emphasize that "status-quo bias is not a mistake--like a calculation error or an error in maximizing" (Samuelson and Zeckhauser, 1988, p.9), but a true behavioral anomaly. While the first two mentioned causes are characteristics of the decision environment, the latter two causes are rooted in the individual. In our empirical analysis, we will focus on the individual drivers of the status-quo bias, and hence also put them into the focus of the following theoretical considerations.

   In the economic literature, the status-quo bias is directly associated with loss aversion and the endowment effect (Kahneman et al., 1991; Tversky and Kahneman, 1991). Thaler (1980, p.44), describes the endowment effect as an "underweighting of opportunity costs" as compared to out-of-pocket cost, which can be explained by loss aversion as established by prospect theory (Kahneman and Tversky, 1979). If the opportunity cost of remaining in the current situation are perceived as foregone gains and the out-of-pocket cost of changing the situation are perceived as a loss, then, according to Thaler (1980), prospect theory's value function implies that the foregone gains (opportunity costs of remaining in the reference state) will be weighted less than the losses (out-of-pocket costs). As a consequence, goods in the current endowment of the individual are valued more highly than goods that are not part of the individual's current endowment. The endowment effect can explain, for example, the differences in buying and selling prices, i.e. differences in willingness to pay (WTP) and willingness to accept (WTA), in various market situations (Kahneman and Tversky, 1984).

   Kahneman et al. (1991) consider both the endowment effect and status-quo bias as a manifestation of loss aversion, or--in other words--loss aversion to be the underlying cause for the status-quo bias and the endowment effect. Loss aversion (in riskless choices) has been conceptualized by Tversky and Kahneman (1991) as an extension of their work on the analysis of choice under uncertainty (Kahneman and Tversky, 1979, 1984), in which they propose that individuals evaluate the outcomes of risky prospects by an asymmetric, S-shaped value function, whose curvature is accounting for reference dependence (definition of gains and losses relative to a reference point, i.e. the origin), loss...