Consumer Preferences for Solar Energy: A Choice Experiment Study.

• Author: Thacher, Jennifer A.

1. INTRODUCTION

   As the urgency of climate change rises, electricity generation in the U.S. is rapidly moving away from coal-fired generation to more environmentally-friendly fossil fuels and, increasingly, towards renewables. The move toward renewables is due to several factors, including cost competitiveness, consumer preferences, and state and federal policies, such as federal tax incentives, renewable portfolio standards (RPSs), and state-level subsidies for behind-the-meter solar. There are costs assoaciated with these subsidies and RPS, which will eventually trickle down to taxpayers and consumers. As such, these costs might affect residents' preferences toward integrating more RE into the system in the future. The question then becomes how much customers are willing to pay for renewable energy (RE). Lastly, consumer preferences may be a key factor in the type of RE that is installed.

   Studies generally do not assess preferences and willingness to pay (WTP) toward a specific RPS goal, but rather evaluate either a generic RE term or its different types (i.e., wind, solar, etc.). Researchers generally find that citizens around the globe have a statistically significant and positive WTP for cleaner electricity. Further, they identify and link heterogeneity in preferences for RE to a number of factors, including energy type, respondents' exposure or proximity to RE, respondents' place of residence (urban/rural), and people's attitudes towards the environment.

   Multiple factors can affect customers' opinions and WTP for RE. There are a plethora of policies and programs currently in place, with high variability in their designs, levels of renewables required, and the number of changes to current standards considered in different locations. However, there is still a lack of understanding of consumer preferences for the level and/or types of renewables, particularly solar energy.

   To fulfill the aforementioned gaps, we conducted a choice experiment survey focusing on preferences for different types of solar energy, an area of investigation that is largely missing within the nonmarket valuation literature. The survey is conducted in New Mexico (NM), a state with an RPS and great potential for renewables, particularly in solar, where it ranks third in the U.S. Focusing on the state's major utility consumer base, our choice experiment considers an increase in the RPS and, specifically, preferences for different types of solar. In addition to gauging households' WTP for a higher level of RPS, we assess households' attitudes towards smart meters (defined below) in NM. Our evaluation considers factors that are expected to be responsible for variation in preferences: distance to the closest solar installation (solar farm and rooftop solar), location (rural versus urban), and environmental worldview. We find that, on average, respondents are supportive of increasing the RPS level, prefer the extra RPS to come from solar farms, and are supportive of smart meter installation. Additionally, rural respondents are significantly more supportive of solar farm improvement than urban respondents. We observe that distance to the nearest solar installation (solar farm and/or rooftop solar) affects preferences toward different types of solar energy. Lastly, we find greater commitment to environmental conservation, as measured by a modified New Ecological Paradigm instrument, leads to higher support for environmental attributes.

2. BACKGROUND

   In 2016, the U.S. contributed approximately 15% of global carbon dioxide emissions (Boden et al., 2017). The electric sector is responsible for the largest share of U.S. emissions (28%) (EPA, 2016). Electricity generation in the U.S. is rapidly moving towards integrating more renewables into the system. Twenty-three percent of electricity generation in the U.S. was from renewable resources in April 2018. (1) The move toward renewables may, in part, be due to renewable portfolio standards (RPS), which are currently required by 29 states (Barbose, 2017).

   RPS policies mandate electric providers to generate a portion of their generation or sale from RE within a certain time frame. The main goal of RPSs policies is to reduce greenhouse-gas emissions and air pollution by reducing the use of fossil fuels, particularly within the electric sector. Subsequently, RPS also may help with reducing water consumption and have impacts on state economies through job creation (Barbose et al., 2016). RPS designs are unique to each state and mainly focus on wind and solar energy. Some RPS has distinct goals solely for solar energy. Among the 29 states with RPS requirements, 18 states have mandated that electric providers within respective states include a minimum amount (carve-out) from solar energy. (2)

   There is a lack of consensus on whether RPSs have a statistically significant and positive effect on RE generation (Carley, 2009; Shrimali et al., 2015; Upton and Snyder, 2017; Wiser et al., 2017; Yin and Powers, 2010), as well as economy and the environment (e.g., Wei and Rose, 2014; Yi, 2015; Wiser et al., 2016; Mamkhezri et al., 2017; Divounguy et al. 2017; Mamkhezri, 2019). (3) Nevertheless, researchers agree RPSs are costly and might directly (e.g., electricity price) or indirectly (e.g., tax) impact citizens (Upton and Snyder, 2017; Palmer and Burtraw, 2005). This may influence taxpayers' opinions on RE in the long-run. Thus, in this research, we assess consumers' preferences towards RE, with a focus on solar energy.

   We are interested in estimating MWTP for two different types of solar generation scales (solar farm and rooftop) for the following reasons. Electric utility companies with required residential solar carve-outs provide subsidies for rooftop solar to encourage their installation, which may impact all customers' monthly electricity bill. Similarly, utility companies with required utility-scale solar carve-outs own and purchase electricity from solar farms, which is also reflected in customers' monthly rates. Further, beyond the cost variable, there might be factors in the actual solar energy generation that customers value. For example, consumers might prefer decentralized solar panels (rooftop solar) over centralized ones (solar farm) due to the belief that solar farms give utilities a monopoly power over solar energy and take away the user's independence associated with owning rooftop solar. Further, they look unpleasant, occupy a large amount of land, kill birds, and interrupt deer migratory paths. On the other hand, rooftop solar requires individual households to make investment decisions, and is less efficient and cost-effective than solar farms. Since customers might be paying higher rates on their monthly electricity bill due to solar energy development and might perceive different types of solar generation scales differently, it becomes relevant to assess whether there is a difference in preferences toward different types of solar energy.

   2.1 Review of literature

   Nonmarket valuation studies, the focus of this research, usually do not specify an exclusive RPS goal to gauge respondents' WTP, but instead assess respondents' opinion and WTP towards requiring more RE in the energy mix. Of those that do specify an RPS level, they find that respondents are willing to pay a positive premium for RPS (Mozumder et al., 2011; Nkansah & Collins, 2018). Numerous empirical studies have commonly found electricity consumers have positive WTP for the move to RE around the globe (e.g., Soon and Ahmad, 2015). (4) Previous research has also found and linked heterogeneity in preferences for RE to several factors including, but not limited to: energy type (e.g., Gracia et al., 2012; Nkansah & Collins, 2018), respondents' exposure or proximity to RE (e.g., Vecchiato and Tempesta, 2015; Mollendorff & Welsch, 2017), respondents' geographic location (urban/rural) (e.g., Bergmann et al., 2008; Yoo, 2011), and people's attitudes towards the environment (e.g., Strazzera et al., 2012; Yoo & Ready, 2014).

   Wind power is currently the most studied energy source within the RE acceptance literature. Researchers find positive WTP the majority of the time (e.g., Nkansah & Collins, 2018; Rehdanz et al., 2017), though sometimes find negative WTP (e.g., Groothuis et al., 2008; Lutzeyer et al., 2018). Scholars link the negative WTP for wind energy to multiple factors, including distance decay effect, which refers to lower WTP the farther away respondents live from a RE development, and vice versa. (5) As noted by Welsch (2016), WTP towards solar energy is one of the most under-studied topics in the field of RE acceptance. Borchers et al. (2007) found that solar energy is the most favored RE technology in comparison to wind, farm methane, biomass, and a generic "green" energy. Likewise, Gracia et al. (2012) showed that solar generated electricity is preferred over wind and biomass. This research is the first to distinguish preferences toward photovoltaic solar (also known as solar farm or utility-scale solar) from residential photovoltaic (also known as rooftop solar) or a combination of the two.

   Consumer WTP for wind energy has been studied the most, particularly with respect to exposure or proximity to RE (e.g., Knapp & Ladenburg, 2015; Gudding et al., 2018). (6) To our knowledge, there are only three peer-reviewed papers that investigate proximity to solar energy. Using a hedonic regression approach, Dastrup et al. (2012) found that rooftop solar add 3.6% to the sale price of a house in California. They related this to financial and moral benefits to the rooftop solar owner, known as "warm glow." Mollendorff & Welsch (2017) measured exposure to solar farm effect on German consumers' well-being and found a statistically significantly negative effect when a solar farm is located in a neighboring postcode district and no significant effect when a solar facility is located within the same postcode district of a respondent...