Applied Energy 324 (2022) 119772

Contents lists available at ScienceDirect

Applied Energy
journal homepage: www.elsevier.com/locate/apenergy

Customer trust in their utility company and interest in household-level

battery storage
Thomas Familia *, Christine Horne
Department of Sociology, Washington State University, 150 Wilson Short Hall, Pullman, WA, USA

H I G H L I G H T S

• Household distrust in utility companies increases interest in battery storage.

• Household beliefs about utility company incompetence increase battery storage interest.
• Household perceptions that battery storage is financially and environmentally beneficial increase interest.

A R T I C L E I N F O A B S T R A C T

Keywords: We seek to contribute to understanding household interest in battery storage, particularly the role of customer
Trust trust in their utility company. Using data from a survey of California households, we examine whether customers’
Battery storage trust in their utility company is associated with their interest in battery storage. We find that customer trust that
Household energy decisions
their utility company acts in their best interest is associated with lower levels of interest. Trust that the utility
Electric utility
company is competent has some effect. Perceptions that battery storage has financial and environmental benefits
and low nonmonetary costs are also associated with higher levels of interest.

1. Introduction the lack of predictability inherent in renewable energy sources. How­

ever, because residential battery storage is relatively new, researchers
The electricity industry in the United States is increasingly incor­ know little about the factors that drive interest in it.
porating renewable energy resources into the electric grid [1]. Primary Our paper seeks to contribute to the understanding of household
sources are wind and solar power; solar panels dominate at the house­ interest in battery storage, particularly the role of customer trust in their
hold level [2,3]. These renewable sources are more variable than fossil utility company. Because existing research highlights the importance of
fuels such as coal and natural gas — the sun does not always shine and costs and benefits in household energy behavior, we also assess their
the wind does not always blow. In addition, renewable sources such as associations with interest in battery storage. We test our hypotheses with
rooftop solar panels installed by households cannot be completely survey data collected from California households in the fall of 2019. The
controlled by utility companies. This lack of predictability and control results are consistent with existing research highlighting consumer
poses major challenges to electric utilities in maintaining a reliable grid. concern with costs and benefits, but also show that customer distrust of
A strategy for addressing these challenges is to store electricity their utility company is associated with greater interest in battery
during times of peak production to use when supplies are low. Because storage. Our findings highlight the potential importance of consumer
demand and supply of electricity must be balanced at all times, and trust for the transition to a greener electricity delivery system and sug­
electricity cannot be stored in the way that other household products gest that utility companies interested in engaging their customers may
can, utility companies have traditionally had few storage options. But need to pay attention to trust.
with the increasing quality of battery storage and the development of
household-level options, there are now opportunities for households to 2. Literature and hypotheses
store electricity produced when the sun is shining for use when supplies
are low or demand is particularly high. Such storage can help address Research on public acceptance of technologies such as nuclear

* Corresponding author.
E-mail addresses: tfamilia@wsu.edu (T. Familia), chorne@wsu.edu (C. Horne).

https://doi.org/10.1016/j.apenergy.2022.119772
Received 10 March 2022; Received in revised form 16 July 2022; Accepted 27 July 2022
Available online 7 August 2022
0306-2619/© 2022 Elsevier Ltd. All rights reserved.
T. Familia and C. Horne Applied Energy 324 (2022) 119772

energy installations or vaccines often focuses on people’s understanding Although we emphasize the potential importance of trust, we also
of risks — that is, how they perceive the probable costs and benefits of expect perceptions of costs and benefits to be relevant. Battery storage
the technology. When people’s assessments differ from those of experts, has potential financial benefits. Customers who have battery storage are
the tendency is to blame this difference on poor assessment capabilities able to store the electricity they buy at low prices to use at times of day
and failures of reasoning and to try to remedy those failures through when prices are high. If they have solar panels, they can also store
education. But decisions may be based on more than just costs and electricity their panels produce to use at night or other times when the
benefits. Sociologists argue that because people are dependent on social sun is not shining, thereby reducing the overall amount of electricity
institutions, their decision-making also reflects their assessments of they buy from their utility company. If they do not currently have solar
those institutions [4]. In particular, people’s trust in relevant institutions panels, they might anticipate such future use. This means that house­
affects the decisions they make [5]. holds may anticipate that battery storage may be able to help them save
Trust is an individual’s expectation that another actor will behave in money on their utility bills. Research on household energy decisions and
a way that benefits (or at least does not harm) the individual [6]. Thus, energy technologies generally shows that for the majority of adopters,
trust occurs in the context of relationships [7,8,9,10]. When an indi­ saving on electricity bills is an important consideration [28]. A related
vidual trusts another actor, they believe that the other intends to act in driver is protecting oneself against future higher costs (e.g., [29,30]).
the individual’s interest (the other cares about the individual’s interests Recent research also shows that some customers are motivated to in­
and not just their own) and that the other is competent [11,12,13]. Thus, crease the value of their homes (which may be the case for both solar and
when a customer trusts their utility company, they believe that the nonsolar households) [31]. Accordingly, we expect that beliefs that
utility company cares about their interests and performs in a competent battery storage will be financially beneficial will increase interest in it.
way. Battery storage may also have environmental consequences [32]. As
Research shows that trust in institutions has declined substantially in noted above, households with battery storage and solar panels can store
the US (e.g., [14]) and other wealthy countries, (e.g., [15]). Low levels electricity to use when the sun is not shining. At times of high demand,
of trust are thought to have a number of negative consequences, solar energy that has been stored can be used instead of electricity from
including reducing effective collective action and lowering people’s coal or natural gas generation plants that must increase capacity to meet
confidence that they can address important problems (e.g., [16]). In the that demand [33]. This means that households can reduce their reliance
energy domain, research shows that distrust in policymakers is associ­ on electricity provided by the utility company (which is likely to be
ated with less acceptance of energy policy [17]. Trust is also important generated by fossil fuels). Conversely, there are negative environmental
for public acceptance of large-scale generation (e.g., [18]). When people impacts related to the battery material lifecycle [34]. Research on en­
distrust installers of electricity generation facilities (such as windfarms ergy technologies often finds that perceived environmental conse­
or energy transmission lines) they are less accepting of those facilities (e. quences affect decision-making (e.g., [35]). And there is some evidence
g., [19,20]). that opinions about the environmental impacts of battery storage vary
Customer trust in companies also has significant implications for [36]. Consistent with existing literature, we predict that:
their engagement with those companies [21]. Trust research shows that
people who trust a company are more likely to engage with it (for H2: Perceptions that battery storage is financially and environmen­
example, by purchasing items from it), whereas those who do not trust a tally beneficial will increase interest.
company will take their business elsewhere (e.g., [22]). In general, when
people do not trust a company, they are likely to want to distance Battery storage also comes with potential costs. The most obvious of
themselves from it. these is financial. In general, people are less likely to adopt technologies
Despite the research on trust in the energy domain and on customer they view as expensive [37]. The cost of installing a battery storage
trust in companies broadly, there is little research on how customer trust system remains high and many see battery storage as not yet financially
in their utility company might affect customer energy decisions. The few viable [38]. Battery storage may also come with nonmonetary costs.
studies that exist show that people who trust their utility company may People may be wary of the difficulties of identifying the right kind of
be more willing to participate in a utility program [6] and to purchase system and getting it installed, and they may be concerned about
home energy management systems [23]. Data suggest, however, that maintenance (e.g., [39]). If people perceive high financial and nonfi­
customers generally have low levels of trust in their electric utilities. For nancial costs, they may be less interested in battery storage.
example, polling data show that people have less trust in electric and gas
utilities than in other industries (e.g., restaurants, technology, auto, H3: Perceptions that battery storage is costly will decrease interest.
farming and agriculture) [24]. Similarly, focus group research shows
consumers placing little trust in power companies [25]. And people tend 3. Methods
to see utility policies and bill increases as self-serving [26].
As described above, trust research suggests that such low levels of To test these hypotheses, we rely on survey data collected from
trust are likely to lead people to want to disengage from a company. But California homeowners in the fall of 2019. The survey covered a range of
customers in the utility company context have little power to do this. questions related to solar energy and battery storage. The Washington
Typically, in the US, people can only purchase electricity from the utility State University Institutional Review Board determined that the study
in their geographic area. They cannot simply leave one utility company was exempt from review (Certificate of Exemption #17222-01); an
for another. They are captive. So, what might they do? One possibility is invitation letter provided recipients with information about the study,
that customers may seek other ways to increase their independence. enabling them to make an informed choice about whether or not to
Research suggests, for example, that people’s interest in rooftop solar participate.
panels is driven, in part, by their interest in becoming more independent
from their utility company by producing at least some of their own 3.1. Sampling and procedures
electricity (e.g., [27]). Investment in battery storage may constitute
another kind of step that customers can take to increase independence. California is a useful site for this research because it is on the leading
The implication is that people who distrust their utility company will be edge of integrating renewable energy sources into the grid and has a
more interested in battery storage. Accordingly, we predict that: sizable share of installed small-scale battery storage — 83 percent of
residential battery storage in the US is located in California [40]. Thus,
H1: Customer trust in their electric utility will be negatively associ­ although penetration of battery storage is still low, California residents
ated with interest in battery storage. may be somewhat more familiar with battery storage than people in

2
T. Familia and C. Horne Applied Energy 324 (2022) 119772

Table 1 Community Survey [44]. For example, 74 percent of the sample has a
Descriptive statistics. bachelor’s degree or higher, whereas only 47 percent of the ACS sample
Variable Mean or Percent (sd) N has a degree. Mean income in the sample is $129,377, compared to
$108,219 in the ACS sample. And 69 percent of the sample is white,
Battery Storage Interest 3.16 1.46 3075
Intent 2.47 1.01 3144 compared to 57 percent in the general population. For details on soci­
Competence: Costs 2.80 0.99 3112 odemographic characteristics see Appendix Table A2.
Competence: Fires 2.67 1.19 3108
Competence: Reliability 3.92 0.93 3130
Competence: Service 3.50 0.89 3110 3.2. Measures
Benefits 3.52 0.82 1933
Nonmonetary Costs 2.83 0.73 1839 The dependent variable is respondents’ interest in residential battery
Monetary Costs 2.72 1.14 2344
storage. We asked participants how interested they were in getting
Env Worry 4.17 0.78 2923
Education 2.14 1.29 2994 household battery storage (1 = not at all; 5 = very) (for exact wording of
Liberal 4.31 1.61 2857 all questions see Appendix Table A3).
Age 59.13 14.12 2903 Our independent variables are trust, benefits, and costs. To assess
White 69 % n/a 2921 trust, we asked questions to get at both the intent and competence
Female 42 % n/a 2996
components of the concept. To assess perceptions of utility company
Income (10 k) 12.94 5.81 2717
Solar Potential 4.34 0.85 3166 intent, we asked participants how much they trusted their utility com­
Have Solar 71 % n/a 3211 pany to act in their best interest (1 = strongly distrust; 5 = strongly
trust). This measure is consistent with existing approaches to studying
trust in institutions (e.g., [45,46]). To assess perceptions of utility
other parts of the country and better able to answer questions about it.
company competence, we focused on four domains, asking how good a
The survey included two samples – households that had solar and
job the respondent’s utility company did at providing a reliable supply
those that did not – as we expect trust, costs, and benefits to be relevant
of electricity, providing good customer service, keeping costs reason­
for both groups. In California, people who install solar are required to
able, and maintaining the grid to prevent wildfires (1 = very bad; 5 =
obtain a permit [41]. Many counties make this permit information
very good).
publicly available on county websites. We scraped county sites for
To measure respondents’ perceptions of financial and environment
permit addresses producing data for 82,439 filings (for a list of counties
benefits we asked a series of questions asking how much they agreed
with permit information see Appendix Table A1). To identify non­
that battery storage could save their household money in the long run,
adopters, we purchased mailing lists for these counties. We then
improve the resale value of their home, reduce their household’s carbon
randomly selected respondents from these two samples. The total
emissions, and reduce their household’s environmental impact (1 =
number of eligible addresses was 26,023 – 12,353 households with solar
strongly disagree; 5 = strongly agree). These items were highly corre­
and 13,670 households without solar.
lated (Cronbach’s alpha = 0.86).
The survey was conducted using the Tailored Design Method [42]
To assess participants’ perceptions of the monetary costs of battery
and used a mail push to web design. Households were initially mailed a
storage, we asked respondents how affordable battery storage was for
letter providing them with a link to the study and inviting them to
them. We recoded this so that affordable = 1 and unaffordable = 5. To
participate. The invitation included a one-dollar pre-incentive. To follow
assess perceptions of nonmonetary costs, we asked them to what extent
up, households were sent a post card reminder. Next, a paper copy of the
they agreed that battery storage was a hassle to install for their house­
questionnaire was mailed to the household. Finally, a last reminder was
holds and difficult to maintain (1 = strongly disagree; 5 = strongly
mailed to non-responders. We obtained a total of 3402 surveys for a 13.1
agree). The two nonmonetary cost measures were correlated (Cron­
percent response rate. This included 2,234 solar adopters (a 18.1 %
bach’s alpha = 0.68).
response rate) and 1168 nonadopters (8.5 % response rate). Although
Because research shows that environmental attitudes may be rele­
our invitation letter stated that we were interested in people’s opinions
vant for household energy decisions, we also included questions asking
on whether they had solar or not, people who did not have solar may
participants about their environmental commitments (1 = strongly
have felt they had little to contribute. In addition, our response rate may
disagree; 5 = strongly agree). These items were also highly correlated
have been affected by the wildfires, outages, and evacuations in Cali­
(Cronbach’s alpha = 0.93).
fornia in the fall of 2019.
We conducted a principal components factor analysis with varimax
Because of our focus on homeowners and our response rate, our
rotation that included the battery benefits, battery costs, and environ­
sample is likely not representative of Californians generally. Instead,
mental items. This yielded three factors – benefits, nonmonetary costs,
participants have characteristics similar to those of solar adopters [43].
and environmental worries. The Eigenvalue for benefits was 2.374, for
Survey participants are more educated, wealthier, and whiter than the
nonmonetary costs 1.386, and for environmental worry was 5.637.
general population in California as determined by the American
Monetary costs did not load with any other factors. The model explained

Table 2
Correlations.
[1] [2] [3] [4] [5] [6] [7] [8] [9]

[1] Battery Storage Interest 1

[2] Intent − 0.18 1
[3] Competence: Costs − 0.15 0.56 1
[4] Competence: Fires − 0.15 0.53 0.47 1
[5] Competence: Reliability − 0.12 0.29 0.39 0.40 1
[6] Competence: Service − 0.15 0.47 0.54 0.48 0.56 1
[7] Benefits 0.48 − 0.05 − 0.05 − 0.09 0.00 − 0.04 1
[8] Nonmon. Costs − 0.24 0.08 0.00 0.08 − 0.02 0.01 − 0.26 1
[9] Monetary Costs − 0.07 − 0.04 − 0.05 0.01 − 0.02 0.01 − 0.06 0.07 1

Note. N = 1283

3
T. Familia and C. Horne Applied Energy 324 (2022) 119772

Table 3 customer service, controlling costs, or managing wildfire risks. How­

OLS regressions showing associations with interest in battery storage. ever, perceptions that the utility did a good job providing a reliable
Model 1 Model 2 Model 3 supply of electricity was negatively associated with interest (in the
models that included battery costs and benefits) (Models 2 and 3). A one
Intent − 0.15** − 0.13** − 0.11*
(0.05) (0.04) (0.04) unit increase in Reliability is associated with a decrease of 0.11 in interest
Competence: Costs − 0.04 − 0.05 − 0.04 (Model 3). The worse respondents thought the utility company was
(0.05) (0.05) (0.05) doing in terms of reliability, the more interested they were in battery
Competence: Fires − 0.06 0.00 0.01 storage.
(0.04) (0.04) (0.04)
Competence: Reliability − 0.05 − 0.09* − 0.11*
Hypothesis 2 predicts that perceptions that battery storage is bene­
(0.05) (0.05) (0.04) ficial will be associated with interest. The statistically significant posi­
Competence: Service − 0.08 − 0.06 − 0.06 tive coefficients for Benefits are consistent with this prediction (Models
(0.06) (0.05) (0.05) 2–3). Our results suggest an increase of 0.76 in interest for every-one
Benefits 0.79*** 0.76***
–
unit increase in Benefits (Model 3).
(0.04) (0.04)
Nonmonetary Costs – − 0.22*** − 0.17*** Hypothesis 3 predicts that perceptions that battery storage is costly
(0.05) (0.05) will reduce interest. We find a negative association between nonmone­
Monetary Costs – − 0.06 − 0.04 tary costs and interest (Models 1–3). A one unit increase in nonmonetary
(0.03) (0.03) costs is associated with a 0.17 decrease in interest in battery storage
Env Worry 0.06
(Model 3). We find no association between monetary costs and interest
– –
(0.06)
Educ – – 0.04 (Models 1–3). Note that although costs do not predict interest, they may
(0.03) still help to explain why people actually install battery storage. That is,
Liberal – – − 0.01 people might express interest in battery storage even though current
(0.03)
costs are high enough that they have no immediate plans to purchase it.
Age – – − 0.01**
(0.00) Their interest may capture a desire to install battery storage sometime in
White – – − 0.08 the future, and the timing of an actual purchase may be associated with
(0.08) costs. Conversely, people may see battery storage as highly affordable
Female – – − 0.09 and yet have no interest in it.
(0.08)
Our control variables have few effects. We find no association be­
Income (10 k) – – 0.01
(0.01) tween environmental worries and interest in battery storage — re­
Solar Potential – – 0.05 spondents with pro-environmental commitments are not more
(0.05) interested in battery storage and liberals are not more interested than
Have Solar 0.57***
– –
conservatives (Model 3). These findings, in conjunction with the find­
(0.08)
Const. 4.45*** 2.40*** 1.71*** ings regarding reliability, suggest that respondents view battery storage
(0.19) (0.30) (0.42) as a strategy for protecting themselves from outages rather than as a tool
R2
0.04 0.28 0.32 for managing household carbon emissions and environmental impact.
Note. N = 1283. Unstandardized coefficients; standard errors in parentheses; *** Households that have solar are more interested in battery storage than
p <.001, ** p <.01, * p <.05. those without. This makes sense, as storage enables households to make
better use of the electricity produced by their solar panels. Older people
72.3 % of the cumulative variance. For the complete results and item are less interested in battery storage than younger people, perhaps
loadings see Appendix Table A4. reflecting age-related comfort with new technologies (e.g., [50]). Other
Finally, sociodemographic characteristics of households have been sociodemographic characteristics (respondent education, gender, race,
shown to be relevant for understanding perceptions of and interest in and income) have no effects. Given that our sample is not representative
residential energy technologies [37,47]. Therefore, we collected the of Californians generally, the results regarding the effects of socio­
basic sociodemographic information described above. We also gathered demographic characteristics should not be interpreted as generalizing to
data on whether each household had rooftop solar. the larger population.

4. Results 5. Discussion

The analyses reported here exclude respondents who had don’t know The results are partially consistent with our predictions. Respondent
responses or missing data (for alternative and exploratory analyses see beliefs that their utility companies have positive intent toward cus­
Appendix B Tables B1-B3). Table 1 reports mean responses to the in­ tomers are positively associated with interest in battery storage. How­
dependent and dependent variable questions. Table 2 reports the cor­ ever, their perceptions that the utility company is competent, for the
relations between the theoretically relevant variables. None of the most part, are not associated with interest. Only beliefs that the utility
predictors in our models showed variance inflation factors higher than company does a bad job providing a reliable supply of electricity in­
1.94, suggesting that multicollinearity is not a problem for the models crease interest. We interpret the trust results as suggesting that when
we fit [48,49]. people believe that their utility company does not care about customers,
To test our hypotheses, we conducted regression analyses with in­ they are more interested in distancing themselves from it. If they do not
terest in battery storage as the dependent variable (see Table 3). Hy­ believe that their electricity supply is reliable and the utility company
pothesis 1 predicts that people who trust their utility company will be does not do a good job restoring power, they may be more interested in
less interested in battery storage. To test this hypothesis we use our battery storage in order to maintain electricity for their household
intent and competence measures of trust. Consistent with our hypoth­ during power outages. Perceptions that battery storage is beneficial in­
esis, we find a statistically significant negative association between crease interest; perceptions that getting a battery system is difficult and
Intent and interest in battery storage (see the Intent coefficient Models that maintenance is a hassle are associated with lower interest.
1–3). Focusing on model 3, the results suggest a one unit increase in The trust literature tends to highlight the positive collective out­
Intent to be associated with a 0.11 decrease in interest. We find no effect comes of trust [16]. Here we show that the consequences of trust may be
of perceptions that the utility company was doing a good job with complicated – distrust increases interest in behaviors (installing battery
storage) that may have positive environmental consequences that

4
T. Familia and C. Horne Applied Energy 324 (2022) 119772

benefit many people. In order for batteries to benefit the collective, factors drive trust in different kinds of entities [56,57] and from whom
however, customers will need more trust in their utility company rather customers prefer to purchase energy technologies [58]. We acknowl­
than less. Further, we find that, of the intent and competence dimensions edge that trust in utility companies, although significant, is only one
of trust, intent has the most consistent effects. Our results suggest a more dimension relevant to battery storage adoption. Because battery storage
complicated picture of the relation between trust and the collective good is just one element of the energy delivery system, consumer trust in other
than is sometimes assumed. relevant actors may also be relevant. For example, in the case of
As a practical matter, the fact that distrust in utility companies is microgrid implementation, trust in neighbors and in the quality of data
associated with greater interest in household battery storage is a po­ protection may also be important [59].
tential problem. As the electric grid incorporates more distributed Although the primary purpose of the study was to examine the role of
renewable energy (such as customer solar panels) and storage (house­ trust, we also asked questions about the benefits and costs of battery
hold solar panels), utility companies will no longer simply sell electricity storage. As rates of uptake increase, people’s understanding of these
to consumers. Instead, consumers will also be producers and storers of benefits and costs will likely grow and shift. Future studies may want to
electricity (e.g., [51]). The fact that renewable energy is less predictable assess customers’ knowledge regarding a broader range of battery
than fossil fuels (the sun does not always shine) further increases the characteristics.
challenges of managing the grid and ensuring that demand and supply Our survey was disseminated at one point of time. The data are cross-
are balanced. Household battery storage is a lever for addressing im­ sectional and do not provide causal evidence. Future research could use
balances. Residential batteries can be locally aggregated and dispatched experimental designs to assess factors that affect trust in industry actors
in real-time [52]. But in order for utility companies to coordinate supply and how trust affects interest in energy technologies. Longitudinal
and demand, they will need to work with customers, particularly those research may also give some purchase on causal associations. Our study
who produce electricity (through rooftop solar panels) and store it also measures interest in battery storage, not actual adoption.
(through battery storage). This will be difficult if households do not trust
their utility company. If households see solar panels and battery storage
as a strategy for becoming more independent from their utility company, 5.2. Conclusion
they are unlikely to want to coordinate more closely with it. And without
such coordination, household production and storage of electricity may We find that people who trust that their utility company has positive
create problems for the electricity delivery system (and other customers) intent towards customers are less interested in battery storage; those
rather than ameliorate them. who trust that their utility company does a good job providing a reliable
supply of electricity may also be less interested. Utility companies have
historically invested substantially in reliability, and more recently have
5.1. Limitations
emphasized customer service. Moving forward, demonstrating the
concern of industry actors for the well-being of customers may also be
Because of the response rate and our focus on homeowners, the re­
important. Our research contributes to the understanding of why
sults cannot be generalized to Californians broadly. Further, it is possible
households might be interested in battery storage and highlights the
that California is unique because of its wildfire history and high elec­
potential importance of customer trust in utility companies for the en­
tricity prices. We expect that levels of customer trust in their utility
ergy transition.
company may vary across states depending on history — extreme
weather events, reports of utility malfeasance, high electricity costs, and
so forth. Consumer trust may also vary with the characteristics of the CRediT authorship contribution statement
household. Therefore, we do not expect levels of distrust to be the same
across locations. However, it is worth keeping in mind that over the last Thomas Familia: Conceptualization, Writing – original draft,
decade many regions of the United States have experienced extreme Methodology, Formal analysis. Christine Horne: Funding acquisition,
weather events causing extended power outages. From the Sandy Hur­ Project administration, Investigation, Writing – review & editing,
ricane that devastated the Northeast in 2012, the Irma and Dorian Supervision.
hurricanes that heavily impacted Florida and the Eastern coastline in
2017 and 2019, to the cold freeze in Texas in 2021, many states have
Declaration of Competing Interest
faced an increase in large-scale natural disasters and accompanying
outages. It is likely that customer distrust in electric utilities is an issue in
The authors declare that they have no known competing financial
these states [53–55]. Whatever the level of distrust in a state, we expect
interests or personal relationships that could have appeared to influence
the prediction regarding the associations between customer perceptions
the work reported in this paper.
that their utility company does not care about their interests and
customer interest in battery storage to apply broadly. Future research
Data availability
should assess the associations between trust and interest in different
contexts, while also considering the practicality of household produc­
Data will be made available on request.
tion of renewable energy (such as solar). Future research should also
assess the effects of the intent and competence components of trust. It
may be that regions with different histories of utility competence react Acknowledgment
differently to these indicators. Future research could also examine
whether the ownership structure of utility companies matters. Does trust This material is based upon work supported by the U.S. Department
vary depending on whether the utility is investor owned (and profit of Energy, Office of International Affairs and Office of Electricity under
driven) or operated by a municipality or cooperative? Award Number DE-IA0000025.
We have focused here on utility companies because, at present, they
are the key point of intersection between consumers and the electricity Appendix A
delivery system. But moving forward, it is possible that other kinds of
organizations will matter. It may be important to know more about what

5
T. Familia and C. Horne Applied Energy 324 (2022) 119772

Table A1
California rooftop solar record availability.
County Record Availability County Record Availability

Alameda Yes Orange Yes

Alpine No Placer Yes
Amador No Plumas No
Butte Yes Riverside Yes
Calaveras No Sacramento Yes
Colusa No San Benito No
Contra Costa Yes San Bernardino Yes
Del Norte No San Diego Yes
El Dorado Yes San Francisco Yes
Fresno No San Joaquin Yes
Glenn No San Luis Obispo Yes
Humboldt Yes San Mateo Yes
Imperial No Santa Barbara Yes
Inyo No Santa Clara Yes
Kern Yes Santa Cruz Yes
Kings No Shasta Yes
Lake No Sierra No
Lassen No Siskiyou No
Los Angeles Yes Solano Yes
Madera No Sonoma Yes
Marin Yes Stanislaus Yes
Mariposa No Sutter Yes
Mendocino Yes Tehama No
Merced Yes Trinity No
Modoc No Tulare Yes
Monterey Yes Tuolumne Yes
Mono No Ventura Yes
Napa Yes Yolo Yes
Nevada Yes Yuba No

Table A2
Descriptive statistics.
Solar Non-adopters Solar Adopters

Variable Mean or Percent (sd) N Mean or Percent (sd) N

Battery Storage Interest 2.59 1.44 897 3.40 1.40 2178

Intent 2.61 1.00 927 2.41 1.01 2217
Competence: Cost 3.01 0.99 918 2.72 0.98 2194
Competence: Fires 2.84 1.18 916 2.60 1.18 2192
Competence: Reliability 3.97 0.90 923 3.90 0.94 2207
Competence: Service 3.58 0.90 920 3.47 0.89 2190
Benefits 3.38 0.84 494 3.57 0.81 1439
Nonmonetary Costs 2.97 0.74 490 2.78 0.72 1349
Monetary Costs 2.76 1.12 605 2.71 1.15 1739
Env Worry 4.15 0.76 857 4.19 0.79 2066
Education 2.09 1.28 873 2.16 1.29 2121
Liberal 4.34 1.54 819 4.29 1.63 2038
Age 58.30 14.83 841 59.47 13.81 2062
White 64 % n/a 858 71 % n/a 2063
Female 45 % n/a 877 41 % n/a 2119
Income (10 k) 11.98 6.00 775 13.32 5.69 1942
Solar Potential 3.93 1.02 922 4.51 0.70 2244
Have Solar 0% n/a 931 100 % n/a 2280

6
T. Familia and C. Horne Applied Energy 324 (2022) 119772

Table A3
Survey items.
Variable Survey Question / Statement Response Categories Recodes

Battery Q34 (If NO to Q33) On a 1 = not at all

Interest scale from 1 to 5, how 2
interested are you in 3
getting household battery 4
storage? 5 = Very
Intent Q23 How much do you trust 1 = strongly distrust
your electric utility 2 = distrust
company to act in your 3 = neither distrust
best interests? nor trust
4 = trust
5 = strongly trust
Competence: Q27A In your experience, how 1 = very bad
Reliability bad or good a job does 2 = bad
your electric utility 3 = average
company do in the 4 = good
following areas: Provide 5 = very good
a reliable supply of
electricity
Competence: Q27B In your experience, how 1 = very bad
Service bad or good a job does 2 = bad
your electric utility 3 = average
company do in the 4 = good
following areas: 5 = very good
Providing good
customer service
Competence: Q27C In your experience, how 1 = very bad
Costs bad or good a job does 2 = bad
your electric utility 3 = average
company do in the 4 = good
following areas: 5 = very good
Providing electricity at
a reasonable cost
Competence: Q27D In your experience, how 1 = very bad
Fires bad or good a job does 2 = bad
your electric utility 3 = average
company do in the 4 = good
following areas: 5 = very good
Maintaining the
electricity system
to prevent fires
Battery Q37E Storage batteries can 1 = strongly disagree
Env reduce my household’s 2 = disagree
(Benefits environmental impact 3 = neither disagree
Factor) nor agree
4 = agree
5 = strongly agree
6 = don’t know
Battery Q37F Storage batteries can 1 = strongly disagree
Climate reduce carbon 2 = disagree
(Benefits emissions for my 3 = neither disagree
Factor) residence nor agree
4 = agree
5 = strongly agree
6 = don’t know
Battery Q37G Storage batteries can 1 = strongly disagree
SavMon save money for my 2 = disagree
(Benefits household in the long 3 = neither disagree
Factor) run nor agree
4 = agree
5 = strongly agree
6 = don’t know
Battery Q37I Storage batteries 1 = strongly disagree
Resale improve the resale 2 = disagree
(Benefits value of my home 3 = neither disagree
Factor) nor agree
4 = agree
5 = strongly agree
6 = don’t know
Battery Q37B Storage batteries 1 = strongly disagree
Maint require a lot of 2 = disagree
(Non- maintenance 3 = neither disagree
monetary nor agree
Cost 4 = agree
Factor) 5 = strongly agree
(continued on next page)

7
T. Familia and C. Horne Applied Energy 324 (2022) 119772

Table A3 (continued )
Variable Survey Question / Statement Response Categories Recodes

6 = don’t know
Battery Q37C Household battery 1 = strongly disagree
Hassle storage is hassle for my 2 = disagree
(Non- household to install 3 = neither disagree
monetary nor agree
Cost 4 = agree
Factor) 5 = strongly agree
6 = don’t know
Monetary Q37A My household can 1 = strongly disagree Original question
Costs afford battery storage 2 = disagree is reverse coded
3 = neither disagree for this analysis
nor agree
4 = agree
5 = strongly agree
6 = don’t know
Care Q49A I care about conserving 1 = strongly disagree
conserve nature 2 = disagree
(Env 3 = neither disagree
Worry nor agree
Factor) 4 = agree
5 = strongly agree
Care Q49C It is important to me to 1 = strongly disagree
LocEnv take care of the 2 = disagree
(Env environment in my 3 = neither disagree
Worry local community nor agree
Factor) 4 = agree
5 = strongly agree
Care Q49F It is important to me to 1 = strongly disagree
EnvGlobe protect the environment 2 = disagree
(Env for people around the 3 = neither disagree
Worry world nor agree
Factor) 4 = agree
5 = strongly agree
Care Q49G It is important to me to 1 = strongly disagree
EnvFuture protect the environment 2 = disagree
(Env for future generations 3 = neither disagree
Worry nor agree
Factor) 4 = agree
5 = strongly agree
Care Q49I I am worried about 1 = strongly disagree
Climate climate change 2 = disagree
(Env 3 = neither disagree
Worry nor agree
Factor) 4 = agree
5 = strongly agree
Care Q49J I am worried about the 1 = strongly disagree
Climate impacts of climate 2 = disagree
Comm change in my 3 = neither disagree
(Env community nor agree
Worry 4 = agree
Factor) 5 = strongly agree
Care Q49K I am worried about the 1 = strongly disagree
Climate impacts of climate 2 = disagree
Globe change around the world 3 = neither disagree
(Env nor agree
Worry 4 = agree
Factor) 5 = strongly agree
Educ Q52 What is the highest level 1 = less than hs Recoded to 5
of school completed by 2 = hs grad or ged categories:
anyone in the household? 3 = some college, 0 = 1,2,3
no degree
4 = ass degree 1=4
5 = bach degree 2=5
6 = masters degree 3=6
7 = prof or 4=7
doc degree
Liberal Q58 What categories best 1 = extremely 1 = ext cons
describes your political liberal
views? 2 = liberal 2 = cons
3 = slightly liberal 3 = slightly cons
4 = moderate 4 = moderate
5 = slightly 5 = slightly lib
conservative
6 = conservative 6 = liberal
7 = extremely 7 = ext lib
conservative
(continued on next page)

8
T. Familia and C. Horne Applied Energy 324 (2022) 119772

Table A3 (continued )
Variable Survey Question / Statement Response Categories Recodes

Age Q51 What year were you Recoded to age

born? of respondent
White Q53 Which of the following Spanish, Hispanic, white = 1; other = 0
-Q54 categories describe you? or Latino/Latina
Please check all that American Indian
apply. Asian
Black or African
American
Native Hawaiian
or other Pacific
Islander
White
Other (please
specify)
Gender Q50 What is your gender? 1 = male Recoded:
2 = female 0 = male, 1 = female,
3 = other 3 dropped
Income Q57 Please estimate your 1 = less than 25,000 1 = 12,500
_10k total household income 2 = 25000–34999 2 = 30,000
last year before taxes 3 = 35000–49999 3 = 42,000
4 = 50000–74999 4 = 62,500
5 = 75000–99999 5 = 87,500
6 = 100000–149999 6 = 125,000
7 = 150000–199999 7 = 175,000
8 = 200000 or more 8 = 200,000
Income_10k=
Income / 10,000
Solar Q14 My roof gets enough sun 1 = strongly disagree
Potential for solar panels to be 2 = disagree
effective 3 = neither
agree nor disagree
4 = agree
5 = strongly agree
Have Q04 To the best of your yes = 1 Recoded:
Solar knowledge, does your no = 2 0 = no, 1 = yes
residence have any solar
panels on this property?

Table A4
Factor Analysis.
Item Env Worry Benefits Non-monetary Costs

I care about conserving nature 0.766 0.049 − 0.199

It is important to me to take care of the environment in my local community 0.758 0.062 − 0.189
It is important to me to protect the env for people around the world 0.841 0.125 − 0.111
It is important to me to protect the env for future generations 0.830 0.075 − 0.146
I am worried about climate change 0.883 0.155 0.007
I am worried about the impacts of climate change in my community 0.874 0.175 0.034
I am worried about the impacts of climate change around the world 0.889 0.148 0.016
Storage Batteries can reduce my household’s env impact 0.161 0.875 − 0.082
Storage Batteries can reduce carbon emissions for my residence 0.178 0.855 − 0.078
Storage Batteries can save money for my household in the long run 0.056 0.829 − 0.028
Storage Batteries can improve the resale value of my home 0.151 0.677 − 0.251
Storage Batteries require a lot of maintenance − 0.137 − 0.075 0.851
Battery Storage is a hassle for my household to install − 0.022 − 0.152 0.847
Eigenvalue, 5.637 2.374 1.386
% variance explained 43.36 % 18.26 % 10.66 %

Note. Principal components analysis, varimax rotation. We initially ran this analysis including monetary costs. Because nonmonetary cost does not load with other
factors, we reran the analysis excluding it.

9
T. Familia and C. Horne Applied Energy 324 (2022) 119772

Appendix B We also conducted analyses in which we allowed the N for each model to
vary depending on the responses for each model (see Appendix
As noted, the analyses reported in the paper include only re­ Table B1). This means that as we added variables to the model (and had
spondents for whom there were no missing data or don’t know responses. more missing data and don’t knows), the N drops. We find substantially

Table B1
OLS Regression showing associations with battery storage interest.
Model 1 Model 2 Model 3

Intent − 0.16*** − 0.13** − 0.11*

(0.03) (0.04) (0.04)
Competence: Costs − 0.05 − 0.05 − 0.04
(0.04) (0.04) (0.05)
Competence: Fires − 0.09** − 0.02 0.01
(0.03) (0.03) (0.04)
Competence: Reliability − 0.04 − 0.08 − 0.11*
(0.04) (0.04) (0.04)
Competence: Service − 0.04 − 0.05 − 0.06
(0.04) (0.05) (0.05)
Benefits – 0.78*** 0.76***
(0.04) (0.04)
Nonmonetary Costs – − 0.21*** − 0.17***
(0.04) (0.05)
Monetary Costs – − 0.06* − 0.04
(0.03) (0.03)
Env Worry – – 0.06
(0.06)
Educ – – 0.04
(0.03)
Liberal – – − 0.01
(0.03)
Age – – − 0.01**
(0.00)
White – – − 0.08
(0.08)
Female – – − 0.09
(0.08)
Income (10 k) – – 0.01
(0.01)
Solar Potential – – 0.05
(0.05)
Have Solar – – 0.57***
(0.08)
Const. 4.22*** 2.36*** 1.71***
(0.12) (0.27) (0.42)
Obs 2998 1571 1283
R2 0.04 0.27 0.32

Note. Unstandardized coefficients; standard errors in parentheses; *** p <.001, ** p <.01, * p <.05.

Table B2
Missing data analysis - Comparisons of Means Across Full and missing sociodemographic characteristics.
Battery storage non-adopters

Sig mean change: sociodem variable smallest p-value:

Battery Storage Interest None 0.804

Intent None 0.915
Competence: Costs None 0.909
Competence: Fires None 0.897
Competence: Reliability None 0.957
Competence: Service None 0.947
Benefits Female 0.000
Nonmonetary Costs Female 0.000
Monetary Costs Female 0.000

Note. t-test comparing the factors of interest in the full sample to the sample when dropping observations with unique missing values in the following sociodemographic
variables: Age, Female (=1), Educ, White (=1), Income, Solar Potential, and Liberal. Differences are statistically significant if p-value less than 0.05.

10
T. Familia and C. Horne Applied Energy 324 (2022) 119772

Table B3
Missing data summary by gender.
Female Male

Factor Variable Missing % Missing %

Battery Storage Interest 70 5.4 % 101 5.6 %

Benefits Battery Saves Money 500 38.3 % 344 19.1 %
Benefits Increases Home Resale 505 38.7 % 352 19.5 %
Benefits Battery Env Benefit 480 36.8 % 295 16.3 %
Benefits Battery Climate Benefit 509 39.0 % 296 16.4 %
Nonmonetary Costs Battery Maintenance 661 50.7 % 499 27.7 %
Nonmonetary Costs Battery Hassle 587 45.0 % 424 23.5 %
Monetary Costs 442 33.9 % 280 15.5 %

Note. Female Respondents = 1305, Male Respondents = 1805.

similar results for our variables of interest. The exceptions are that in [12] Hardin R. Trust and Trustworthiness. New York: Russel Sage Foundation; 2002.
[13] Walker J, Ostrom E. Trust and Reciprocity as Foundations for Cooperation. In:
Model 1, Utility Fires is statistically significant, and in Model 2, Utility
Cook K, Levi M, Hardin R, editors. Whom Can We Trust: How Groups, Networks,
Reliability is not statistically significant and Monetary Costs is statistically and Institutions Make Trust Possible. New York: Russell Sage Foundation; 2009.
significant. These differences suggest some variation due to don’t know [14] Edelman. 21st Annual Trust Barometer. Washington, DC; 2021. Retrieved August
and missing responses. But our main finding regarding the intent 4, 2021 [https://www.edelman.com/trust/2021-trust-barometer].
[15] Perry Jonathan. Trust in Public Institutions: Trends and Implications for Economic
component of trust persists. However we specify our models, trust as Security. New York: United Nations Department of Social Affairs; 2021. Accessed
assessed in terms of utility intent decreases interest in battery storage. May 23, 2022. [https://www.un.org/development/desa/dspd/2021/07/trust
We also looked at the number of don’t knows and nonresponses for -public-institutions/].
[16] Putnam RD, Leonardi R, Nanetti R. Making Democracy Work: Civic Traditions in
each variable. The highest levels of don’t know responses are for the Modern Italy. Princeton: Princeton University Press; 1993.
battery questions [Q37A, Q37B, Q37C, Q37E, Q37F, Q37G, & Q37I]. [17] Becker S, Demski C, Evensen D, Pidgeon N. Of Profits, Transparency, and
Women were more likely than men to respond don’t know). The number Responsibility: Public Views on Financing Energy System Change in Great Britain.
Energy Res Social Sci 2019;55:236–46. https://doi.org/10.1016/j.
and distribution of don’t knows is not surprising given that household erss.2019.05.013.
battery storage is new and has very little penetration [60–62]. [18] Ryu Y, Kim S, Kim S. Does Trust Matter? Analyzing the Impact of Trust on the
We had relatively high numbers of nonresponses to some of our Perceived Risk and Acceptance of Nuclear Power Energy. Sustainability 2018;10
(3). https://doi.org/10.3390/su10030758.
sociodemographic questions. Missing data for Income (16 %) and Liberal [19] Devine-Wright P. Explaining ‘NIMBY’ Objections to a Power Line: The Role of
(12 %) were high; missing data for Age (10 %) and White (10 %) were Personal, Place Attachment and Project-Related Factors. Environ Behavior 2013;45
also somewhat high. We compared the means of our theoretical vari­ (6):761–81. https://doi.org/10.1177/0013916512440435.
[20] Jami AA, Walsh PR. From Consultation to Collaboration: A Participatory
ables of interest (Intent, Benefits, Nonmonetary Costs, and Monetary
Framework for Positive Community Engagement with Wind Energy Projects in
Costs) for the full sample compared to the samples that dropped cases Ontario, Canada. Energy Res Social Sci 2017;27:14–24. https://doi.org/10.1016/j.
based on these missing sociodemographic responses (examining one erss.2017.02.007.
sociodemographic variable at a time) (see Appendix Table B2). We find [21] Koirala BP, Araghi Y, Kroesen M, Ghorbani A, Hakvoort RA, Herder PM. Trust,
Awareness, and Independence: Insights from a Socio-Psychological Factor Analysis
that means are generally consistent. The only statistically significant of Citizen Knowledge and Participation in Community Energy Systems. Energy Res
differences are that mean responses to the battery benefits and costs Social Sci 2018;38(February):33–40. https://doi.org/10.1016/j.erss.2018.01.009.
questions differ across gender (see Appendix Table B2). Women were [22] Schelly C, Letzelter JC. Examining the Key Drivers of Residential Solar Adoption in
Upstate New York. Sustainability 2020;12:2552. https://doi.org/10.3390/
much more likely than men to respond to the battery questions with su12062552.
don’t know (see Appendix Table B3). [23] Washizu A, Nakano S, Ishii H, Hayashi Y. Willingness to Pay for Home Energy
Management Systems: A Survey in New York and Tokyo. Sustainability
(Switzerland) 2019;11(17):1–20. https://doi.org/10.3390/su11174790.
References [24] Saad Lydia. Restaurants Again Voted Most Popular U.S. Industry. Gallup, August
15; 2016. Retrieved August 21, 2020 [https://news.gallup.com/poll/194570/rest
[1] EIA, US. Solar Power Will Account for Nearly Half of New U.S. Electric Generating aurants-again-voted-popular-industry.aspx].
Capacity in 2022. Long Island City, NY; 2022a. Retrieved May 19, 2021 [htt [25] Goulden M, Bedwell B, Rennick-Egglestone S, Rodden T, Spence A. Smart Grids,
ps://www.eia.gov/todayinenergy/detail.php?id=50818]. Smart Users? The Role of the User in Demand Side Management. Energy Res Social
[2] EIA, US. Monthly Energy Review, March 2022. Long Island City, NY; 2022b. Sci 2014;2:21–9. https://doi.org/10.1016/j.erss.2014.04.008.
Retrieved May 19, 2021 [https://www.eia.gov/totalenergy/data/monthly/archive [26] Palm J, Tengvard M. Motives for and Barriers to Household Adoption of Small-
/00352203.pdf]. Scale Production of Electricity: Examples from Sweden. Sustain: Sci, Pract, Policy
[3] Feldman, David, Robert Margolis. Q2/Q3 2020 Solar Industry Update. Golden, CO: 2011;7(1):6–15. https://doi.org/10.1080/15487733.2011.11908061.
National Renewable Energy Laboratory; 2020. Retrieved August 4, 2021 [https [27] Kalkbrenner, Bernhard J. Residential vs. Community Battery Storage Systems –
://www.nrel.gov/docs/fy21osti/78625.pdf]. Consumer Preferences in Germany. Energy Policy 2019;129 (February 2018):
[4] Freudenburg WR. Risk and Recreancy: Weber, the Division of Labor, and the 1355–63. doi: 10.1016/j.enpol.2019.03.041.
Rationality of Risk Perceptions. Soc Forces 1993;71(4):909–32. https://doi.org/ [28] Vasseur V, Kemp R. The Adoption of PV in the Netherlands: A Statistical Analysis of
10.1093/sf/71.4.909. Adoption Factors. Renew Sustain Energy Rev 2015;41:483–94. https://doi.org/
[5] Büscher C, Sumpf P. ‘Trust’ and ‘Confidence’ as Socio-Technical Problems in the 10.1016/j.rser.2014.08.020.
Transformation of Energy Systems. Energy, Sustain Soc 2015;5(1):1–13. https:// [29] Balcombe P, Rigby D, Azapagic A. Motivations and Barriers Associated with
doi.org/10.1186/s13705-015-0063-7. Adopting Microgeneration Energy Technologies in the UK. Renew Sustain Energy
[6] Stenner K, Frederiks ER, Hobman EV, Cook S. Willingness to Participate in Direct Rev 2013;22:655–66. https://doi.org/10.1016/j.rser.2013.02.012.
Load Control: The Role of Consumer Distrust. Appl Energy 2017;189:76–88. [30] Wittenberg I, Matthies E. Solar Policy and Practice in Germany: How Do
https://doi.org/10.1016/j.apenergy.2016.10.099. Residential Households with Solar Panels Use Electricity? Energy Res Social Sci
[7] Fehr E. On the Economics and Biology of Trust. J Europ Econ Assoc 2009;7(2–3): 2016;21:199–211. https://doi.org/10.1016/j.erss.2016.07.008.
235–66. https://doi.org/10.1162/JEEA.2009.7.2-3.235. [31] Karjalainen S, Ahvenniemi H. Pleasure Is the Profit - The Adoption of Solar PV
[8] Gambetta D. Can We Trust Trust? In: Gambetta D, editor. Trust: Making and Systems by Households in Finland. Renewable Energy 2019;133:44–52. https://
Breaking Cooperative Relations. New York: Basil Blackwell; 1988. p. 213–37. doi.org/10.1016/j.renene.2018.10.011.
[9] Levi M, Stoker L. Political Trust and Trustworthiness. Ann Rev Polit Sci 2000;3: [32] Dehghani-Sanij AR, Tharumalingam E, Dusseault MB, Fraser R. Study of Energy
475–507. https://doi.org/10.1146/annurev.polisci.3.1.475. Storage Systems and Environmental Challenges of Batteries. Renewable Sustain
[10] Przepiorka W, Horne C. How Can Consumer Trust in Energy Utilities Be Increased? Energy Rev 2019;104 (November 2018):192–208. doi: 10.1016/j.rser.2019.01.023.
The Effectiveness of Prosocial, Proenvironmental, and Service-Oriented [33] Maeyaert L, Vandevelde L, Döring T. Battery Storage for Ancillary Services in
Investments as Signals of Trustworthiness. Org Environ 2020. https://doi.org/ Smart Distribution Grids. J Storage Mater 2020;30(May):101524. https://doi.org/
10.1177/1086026618803729. 10.1016/j.est.2020.101524.
[11] Coleman JS. Commentary: Social Institutions and Social Theory. Am Soc Rev 1990;
55(3):333–9. https://doi.org/10.2307/2095759.

11
T. Familia and C. Horne Applied Energy 324 (2022) 119772

[34] Le Varlet, Thomas, Oliver Schmidt, Ajay Gambhir, Sheridan Few, Iain Staffell. [49] Neter J, Wasserman W, Kutner MH. Applied Linear Regression Models. Homewood,
Comparative Life Cycle Assessment of Lithium-Ion Battery Chemistries for IL: Irwin; 1989.
Residential Storage. J Energy Storage 2020;28 (September 2019):101230. doi: [50] Czaja SJ, Charness N, Fisk AD, Hertzog C, Nair SN, Rogers WA, et al. Factors
10.1016/j.est.2020.101230. Predicting the Use of Technology: Findings from the Center for Research and
[35] Noppers E, Keizer K, Milovanovic M, Steg L. The Role of Adoption Norms and Education on Aging and Technology (CREATE). Psychol Aging 2006;21(2):333–52.
Perceived Product Attributes in the Adoption of Dutch Electric Vehicles and Smart https://doi.org/10.1037/0882-7974.21.2.333.
Energy Systems. Energy Res Social Sci 2019;57(July):101237. https://doi.org/ [51] Gough M, Santos SF, Javadi M, Castro R, Catalão JPS. Prosumer Flexibility: A
10.1016/j.erss.2019.101237. Comprehensive State-of-the-Art Review and Scientometric Analysis. Energies
[36] Thomas G, Demski C, Pidgeon N. Deliberating the Social Acceptability of Energy 2020;13(11):1–32. https://doi.org/10.3390/en13112710.
Storage in the UK. Energy Policy 2019;133:110908. https://doi.org/10.1016/j. [52] Goyal M, Ghosh A. Microgrids Interconnection to Support Mutually During Any
enpol.2019.110908. Contingency. Sustain Energy Grids Networks 2016;6:100–8. https://doi.org/
[37] Schelly C. Residential Solar Electricity Adoption: What Motivates, and What 10.1016/j.segan.2016.02.006.
Matters? A Case Study of Early Adopters. Energy Res Social Sci 2014;2:183–91. [53] Cuomo, Andrew M. Governor Cuomo Introduces Legislation to Provide Additional
https://doi.org/10.1016/j.erss.2014.01.001. Enforcement Tools to Hold Utilities Accountable. New York, NY: Office of the
[38] Castagneto Gissey G, Subkhankulova D, Dodds PE, Barrett M. Value of Energy Governor; 2020. Retrieved August 4, 2021 [https://www3.dps.ny.gov/pscweb/
Storage Aggregation to the Electricity System. Energy Policy 2019;128:685–96. WebFileRoom.nsf/Web/C3E128512004D7A78525860F006575BB/$File/gov
https://doi.org/10.1016/j.enpol.2019.01.037. cuomo introduces leg to provide addtl enforcement tools to hold utilities
[39] Ambrosio-Albalá, Pepa, Paul Upham, Catherine SE Bale. Purely Ornamental? accountable_102820.pdf?OpenElement].
Public Perceptions of Distributed Energy Storage in the United Kingdom. Energy [54] Pounds, Marcia Heroux. FPL Spent $2.78B on Grid Upgrades, but Customers Doubt
Res Soc Sci. 2019;48(October 2018):139–50. doi: 10.1016/j.erss.2018.09.014. They Got Their Money’s Worth. Sun Sentinel, December 3;. 2017. Retrieved August
[40] EIA, US. Battery Storage in the United States: An Update on Market Trends. Long 15, 2021. [https://www.sun-sentinel.com/business/fl-bz-fpl-grid-upgrades-
Island City, NY; 2021. Retrieved May 19, 2021. [https://www.eia.gov/analysis/st 20171127-story.html].
udies/electricity/batterystorage/pdf/battery_storage_2021.pdf]. [55] Treisman Rachel. New York and Connecticut Call For Probes Into Utility
[41] CA OPR. California Solar Permitting Guidebook. Sacramento, CA: Solar Permitting Companies’ Response To Isaias. NPR. August 5; 2020. Retrieved November 14,
Task Force; 2019. Retrieved April 4, 2021 [https://www.opr.ca.gov/docs/20190 2021 [https://www.npr.org/2020/08/05/899535041/new-york-and-connecticut-
226-Solar_Permitting_Guidebook_4th_Edition.pdf]. call-for-probes-into-utility-companies-response-to-isai].
[42] Dillman D, Smyth J, Christian L. Internet, Phone, Mail, and Mixed-Mode Surveys: [56] Sirdeshmukh D, Singh J, Sabol B. Consumer Trust, Value, and Loyalty in Relational
The Tailored Design Method. 3rd ed. Hoboken, NJ: John Wiley; 2014. Exchanges. J Market 2002;66(1):15–37. https://doi.org/10.1509/
[43] Barbose, Galen L, Sydney Forrester, Naïm R Darghouth, Ben Hoen. Income Trends jmkg.66.1.15.18449.
among U.S. Residential Rooftop Solar Adopters. Berkeley, CA: Lawrence Berkeley [57] Strupeit L, Palm A. Overcoming Barriers to Renewable Energy Diffusion: Business
National Laboratory; 2020. doi: 10.2172/1603637. Models for Customer-Sited Solar Photovoltaics in Japan, Germany and the United
[44] American Community Survey; 2019. ACS 1-Year Estimates Subject Tables, TableID: States. J Cleaner Prod 2016;123:124–36. https://doi.org/10.1016/j.
S2502 (Owner-occupied housing units). Retrieved April 4, 2021 [https://data.ce jclepro.2015.06.120.
nsus.gov/cedsci/table?q=S2502&g=0400000US06&tid=ACSST1Y2019.S2502 [58] Ambrosio-Albalá P, Upham P, Bale CSE, Taylor PG. Exploring Acceptance of
&hidePreview=false]. Decentralised Energy Storage at Household and Neighbourhood Scales: A UK
[45] Kavanagh, Jennifer, Katherine Grace Carman, Maria DeYoreo, Nathan Changler, Survey. Energy Policy 2020;138. https://doi.org/10.1016/j.enpol.2019.111194.
Lynn E. Davis. Drivers of Institutional Trust and Distrust: Exploring Components of [59] Soshinskaya M, Crijns-Graus WHJ, Guerrero JM, Vasquez JC. Microgrids:
Trustworthiness. Rand; 2020. Accessed May 23, 2022 [https://www.rand.org/con Experiences, Barriers and Success Factors. Renew Sustain Energy Rev 2014;40:
tent/dam/rand/pubs/research_reports/RRA100/RRA112-7/RAND_RRA112-7. 659–72. https://doi.org/10.1016/j.rser.2014.07.198.
pdf]. [60] Barbose, Galen, Salma Elmallah, Will Gorman. “Behind-the-Meter Solar+Storage:
[46] OECD. Illustrative Examples of Trust Measures. OECD. Accessed May 23, 2022. Market Data and Trends. Berkeley, CA: Lawrence Berkeley National Laboratory;
[https://read.oecd-ilibrary.org/governance/oecd- 2021. Retrieved August 4, 2021 [https://eta-publications.lbl.gov/sites/defa
guidelines-on-measuring-trust/illustrative-examples- ult/files/ btm_solarstorage_trends_final.pdf].
of-trust-measures_9789264278219-9-en#page1]. [61] Chen Z, Garand JC. On the Gender Gap in Financial Knowledge: Decomposing the
[47] Sommerfeld J, Buys L, Mengersen K, Vine D. Influence of Demographic Variables Effects of Don’t Know and Incorrect Responses. Soc Sci Quart 2018;99(5):1551–71.
on Uptake of Domestic Solar Photovoltaic Technology. Renew Sustain Energy Rev https://doi.org/10.1111/ssqu.12520.
2017;67:315–23. https://doi.org/10.1016/j.rser.2016.09.009. [62] Lizotte MK, Sidman AH. Explaining the Gender Gap in Political Knowledge. Polit
[48] Menard, Scott W. Applied Logistic Regression Analysis. Thousand Oaks, Calif: Sage Gender 2009;5(2):127–51. https://doi.org/10.1017/S1743923X09000130.
Publications; 1995.

12