Citizen Engagement in EU Collective Action Energy Projects

Abstract

Transitioning to a low-carbon economy requires profound changes in the energy system, including digitisation and decentralisation. Collective action energy projects are a promising means of supporting this transformation since they redistribute control of energy resources to the citizens and foster sustainable energy behaviours, as well as potentially enhancing energy justice and democracy. The EU has funded a number of collective action energy R&I projects which combine the use of new technologies and community engagement. We identify 28 such projects and investigate the types of community engagement strategies trialed and the challenges they face. We find that engagement strategies comprise a mixture of shorter term participation and longer term behaviour change strategies. There is a tendency towards behaviour change strategies based on feedback and/or rewards, while other promising strategies like targeting group norms are less common. Overall, projects opt for medium-level participation, while co-creative or more empowering approaches are less common. A lack of consistency in approach and a lack of reported data on social impacts of the projects makes it difficult to determine how effective the engagement strategies (and hence projects themselves) have been. We provide some recommendations for designing and monitoring similar research projects going forward.

1. Introduction

Shifting to a low-carbon economy calls for increasing digitalisation and electrification of the energy system [1]. Combined with renewable power, these transformations can be instrumental for the green energy transition, facilitating energy efficiency improvement in urban areas, sustainable mobility systems, and innovative consumer-centered services. Therefore, the twin challenges of a green and digital transformation must be compatible if Europe wants “to lead the transition to a healthy and a new digital world” [2]. Several EU policy documents have established the central role of smart grid technologies in the energy transition. At the same time, there has been increasing awareness that digital technologies deployment requires active citizen participation [1].

Encouraging consumers to change their behaviour, to reduce consumption, and increase energy efficiency, are important aspects of the energy transition communicated in policy. The European Green Deal acknowledges the importance of the collective dimension of energy use in the energy transition [3]. Empowering and providing local communities/consumers with the tools for greater participation in the energy market is considered as essential to help achieve EU renewable energy targets and for citizens to benefit from internal electricity markets [3,4]. The Clean Energy Package (CEP), for example, establishes a legislative framework where the participation of “active customer” (definition that also includes “a group of jointly acting final customers” (Directive (EU) 2019/944, 2019), article 2, point 8: ‘active customer’ means a final customer, or a group of jointly acting final customers, who consumes or stores electricity generated within its premises located within confined boundaries or, where permitted by a Member State, within other premises, or who sells self-generated electricity or participates in flexibility or energy efficiency schemes, provided that those activities do not constitute its primary commercial or professional activity) and “jointly acting renewable self-consumers” (Directive (EU) 2018/2001, 2018), article 2, point 15: ‘jointly acting renewables self-consumers’ means a group of at least two jointly acting renewables self-consumers in accordance with point (14) who are located in the same building or multi-apartment block) is fostered. Furthermore, the electricity directive (Directive (EU) 2019/944, 2019) [5] and renewable energy directive (Directive (EU) 2018/2001, 2018) [6] define ‘citizen energy community” and “renewable energy community” as a particular way to organise collective actions around a specific energy-related activity. These directives require Member States to facilitate community arrangements, which is a new way of governing the energy system. The ensuing decentralization of decision-making and empowerment of households and communities requires a more human-centric approach to policy-making. This creates opportunities but also challenges in terms of participation and empowerment for citizens. Previously, citizens have been viewed as passive customers or recipients of technologies in the energy system [7,8]. However, recent research builds on the theory that energy-related behaviours exist within particular social and technological contexts. Energy use is rather an ingredient in social practices that co-evolve with developments in energy technologies [9,10].

A greater participation of citizens, via energy-related collective action initiatives, or community energy initiatives is linked to a successful energy transition [7,8,11]. Leveraging on communal motives may serve as a means to strengthen participation in community energy initiatives. It may also promote sustainable behaviour in those who are not necessarily interested in environmental protection [11]. Community energy initiatives are also believed to promote energy justice and democracy [12,13,14] and help mitigate energy poverty [15], thus contributing to a socially just energy transition. Energy communities can also bring a number of benefits from a socio-technical perspective including local value, employment, financial benefits, education, social cohesion, energy autonomy, zero-emission energy, or lower air pollution [16].

Collective actions may be defined as ‘the action taken by a group (either directly or on its behalf through an organization) in pursuit of member’s perceived shared interests’ [17]. Collective energy action initiatives have various organisational or management structures ranging from grassroots organisations to cooperatives [18]. Collective energy actions have now started to emerge in the EU Member States, for example, energy cooperatives, peer-to-peer (P2P) trading and collective self-consumption. These run either within existing legal frameworks, under regulatory exemptions or in the framework of innovation projects. While these developments are promising, it remains to be seen as to their efficacy in promoting sustained citizen engagement, empowerment, inclusion or equity, and as a result, energy justice and democracy.

Apart from financial considerations, energy behaviour (which includes participation in energy communities or collective actions) is shown to be influenced by a myriad of factors. These include socio-demographic factors (age, gender, education, income, location etc.), psychological factors (knowledge, awareness, values, beliefs, attitudes, group membership and social influence, perceived moral responsibility etc.), and situational factors (laws, regulations, available technology, pricing, socio-cultural and socio-economic characteristics) [19]. These factors and their interrelationships are not yet fully understood [20], making it challenging to design appropriate strategies to change or influence behaviour. However, generally, community members who feel engaged likely perceive fewer barriers to local energy projects, thereby easing their implementation [21]. Compared to top-down management arrangements, bottom-up or grassroots processes are linked to innovations designed in line with community needs [22]. Acceptance is also contingent on how fair and democratic the project is perceived to be [23,24].

Not surprisingly, recent research shows that recruitment and sustained engagement of participants can still be challenging [16], due to challenges of inadequate profiling of end users, lack of budget for engagement strategies, unmotivated users, or misunderstandings between projects and users [25]. Further, local governance may not always equate to more democratic governance [26]. In particular, engagement activities targeting vulnerable groups tend to be problematic [13].

Understanding the underlying theory and mechanism for various behavior and participatory interventions may help to design well-targeted strategies. The EU has funded a number of research and innovation (R&I) projects which trial a combination of new technologies, community engagement approaches and business models to foster consumer participation in the energy market. By experimenting with different engagement strategies, R&I projects can play a pivotal role in understanding and addressing key challenges and hence accelerate a consumer-centric energy transition in the EU Member States.

The majority of studies of energy community outcomes do not use formal assessment approaches to assess local impacts explicitly [27], meaning that more complete and, generalisable conclusions about how and why social impacts occur cannot be drawn. Yet, assessing the outcomes of engagement strategies used in community-oriented energy projects, such as the impact on behaviours, both individual and collective, or longer-term participation, is important in providing valuable evidence for future policy formulation.

While previous studies e.g., [25] examine general challenges to engagement in EU R&I collective energy projects or whether energy communities address energy justice concerns like energy poverty [13], to our knowledge there is no in-depth overview of the engagement methods used in EU R&I collective energy projects. The 2021 BRIDGE project [25] highlights the need for such a study, noting that many promising methodologies and engagement strategies that could be used by projects are overlooked and that evaluation methods are inconsistent and applied in a patchy manner across project phases. Hanke et al. [13] also suggest that energy collectives could be further empowered by a greater understanding of the barriers they face and the impact of engagement strategies on different social groups.

In this article we focus on those projects that research and test collective level technological solutions to enhance self-produced and locally generated electricity and to promote energy efficiency and active demand services. The technologies trialled in the projects mainly revolve around collective energy demand side management with the goal of improving local energy management, reducing energy consumption and costs for consumers, improving energy independence and supporting participation in local flexibility markets.

Building on previous JRC research on consumer engagement [28] and community-based approaches in demand side management EU projects [29] and extending the findings presented in [30], this article presents a snapshot of current developments in consumer engagement strategies in EU-funded collective action projects in the energy sector. The aim of the article is to investigate the types of community engagement strategies that have been tested in selected collective action energy projects, to analyse the challenges they face and to identify potential improvements that could be made, including ways to improve the evaluation of these projects through meaningful outputs. We discuss our findings in the light of policy-making in the EU and we identify research gaps and areas for research that can help shape future EU-funded R&I calls for proposals to include a socio-technical perspective to the energy transition. These findings can also help to support the EU Member States in the implementation of the electricity and renewable energy directives and help collective energy initiatives to develop engagement strategies that lead to sustained commitment and lasting behavioural change.

The paper is organized as follows: in Section 2 we introduce an overview of engagement strategies used in the energy field; in Section 3, we present the methodological approach; in Section 4, we present the results of the analysis highlighting the findings in terms of behavior change and participatory strategies. We also refer to some of the projects to give examples of interesting emerging practices. Finally, in Section 4, we discuss our findings and provide recommendations.

2. Engagement Strategies in the Energy Transition

Engagement strategies used in the energy transition are found to serve diverse purposes: communication, creating acceptance or buy-in, motivating (continued) participation or behavioural change, as well as upholding certain justice-related and democratic norms [12]. Commonly used interventions/strategies for engagement in the energy transition include interventions for encouraging pro-environmental behaviour in the energy transition as well as participatory approaches, which sometimes overlap (Figure 1).

It is now increasingly evident that behaviour change is more likely to be created by targeting various underlying psychological factors [31]. Social norms are “the explicit and/or implicit rules, guidelines or behavioural expectations within a group or society that guide what is considered normal and/or desirable” [19]. Social norms have an important influence on people’s willingness to participate in energy communities [32,33].

Various behaviour change interventions/strategies are found in the literature, summarized in Figure 2. Such interventions may take the form of “antecedent strategies” or “consequence strategies” [34]. “Antecedent strategies” target underlying behavioural determinants prior to the behaviour itself. They may include information provision, commitments, goal setting, peer influence, social modelling or the use of social/group norms.

“Consequence strategies” are built on the premise that either positive or negative consequences will influence behaviour. These strategies may include feedback on behaviour, monitoring, comparisons to others, rewards or other incentives, and penalties [20,35,36] or gamification which combines several of these approaches together.

Public participation is a cornerstone of socio-technical transformation in democratic societies. It is hence an essential means of bringing about inclusive, democratic and socially just change in the energy system. It involves communicating with citizens, mobilising them, gaining public acceptance for policies or technologies and for encouraging changes in behaviour [12].

The importance of using participatory approaches as a way to enhance the design and implementation of behaviour change interventions is also increasingly recognised by practitioners [35,36]. There is also clearly some overlap between participatory approaches and other behaviour change interventions. For example focus groups, training sessions and public meetings [37] or gamification e.g., [38] may aim to tap into social norms to encourage pro-environmental behaviour, provide information, or build a sense of community, but may also result in non-instrumental outcomes e.g., increased empowerment, or social learning. However, behaviour change interventions tend to have an instrumental rationale, (i.e., agenda) e.g., for promoting buy-in or acceptance of energy projects or some other desirable behaviour, whereas participatory interventions may have one or more rationales, i.e., instrumental (e.g., increase the social acceptability of projects, reduce conflict or increase legitimacy of decision-making), substantive (i.e., participation can improve the quality of decisions, by e.g., incorporating local knowledge or concerns) or normative (e.g.,: normative ideals such as democracy or inclusion of the least powerful) [39]. Not all behaviour change interventions have the same potential to uphold normative principles of democracy or energy justice: for example, information provision, while acknowledged an important first step of participation (Figure 3), when used alone has been criticised for placing an emphasis on a one-way flow of information and therefore reducing the influence of citizens on decision-making [40].

An oft-used model to describe the different degrees of empowerment in citizen participation in (public) decision-making is Arnstein’s ladder of citizen participation [41], which visualises public participation as ranging from passive involvement of citizens as recipients of information or announcements, or respondents to consultations or surveys, through to active participants or owners of the process itself. Figure 3 shows an adapted version of Arnstein’s ladder, inspired also by [42,43].

3. Materials and Methods

Since 2011, the JRC has kept an inventory of smart grid projects, i.e., projects that aim to efficiently integrate the behaviour and actions of all users connected to the electricity network. The JRC inventory is the most comprehensive database of smart grid projects in Europe, covering almost 15 years of EU R&I funding. Given its focus on projects dealing with the integration of distributed energy resources as well as connected users’ behaviours, it represents a useful tool to explore the evolving role of consumers in the energy system, their participation in the energy market and how they can best contribute to the twin green and digital transition.

We screened the 2021 JRC database of smart grid projects funded under the last two framework programs for research and innovation (FP7 and H2020) [1]. The projects were thoroughly screened via a systematic search using various combinations of the following keywords: “innovation”, “collective”, “self-consumption”, “storage”, “peer to peer”, “energy trading”, “energy sharing”, “community”, “energy cooperative”, “neighbourhood”, “virtual power plant”, “microgrid”, “distributed generation”, “demand management”, “demand response”, “smart grids”. The focus of the search was to identify projects that research and test collective level technological solutions which facilitate self-production and consumption of locally generated electricity as well as energy efficiency and active demand solutions and services.

Search results were selected for further analysis only if they met the following selection criteria:

• The project targets the residential sector, either exclusively or in combination with other sectors. This article focuses on residential consumer participation in energy generation, management and consumption. Projects targeted exclusively at the industrial, commercial and public sector were hence not taken into consideration.
• The project actively involves consumers at collective level through the use of an enabling technology and/or engagement strategy based on a sense of community and of shared values and goals.
• There is enough information publicly available, e.g., in the form of deliverables, reports, research articles or a demonstrator description. Several older and recent projects were screened out as after the closing and before the launch of the project website fewer information is usually available.
• The project states clear research objectives and provides for a monitoring and evaluation framework.

The project selection process was carried out in parallel and independently by two researchers and discrepancies between the researchers’ findings were discussed and resolved. At the end of the screening process, we retained 28 collective action projects out of 407 included in the JRC catalogue. Project documentation was scanned to identify the engagement interventions tested, the challenges encountered, the outcomes achieved, and the lessons learned.

This article is limited to the identification of collective engagement strategies in R&I collective actions projects found in the JRC smart grid project inventory, implying a particular technological focus. Conclusions can only be drawn in relation to this type of project and not all types of collective energy projects that exist in the EU. The analysis is also limited by the amount of documentation available from the projects. These projects often feature multiple, interacting strategies; it is therefore not possible to separate the effects of one intervention from another and hence to evaluate the effectiveness of the adopted engagement strategy based on the results.

4. Results and Discussion

Our analysis presents the strategies used by the projects surveyed to engage participants in collective energy projects. Projects typically combine the use of various strategies, which include interventions aiming to change individual or group behaviour towards sustainable energy practices, e.g., via use of a technology, or interventions to stimulate participation in the project in some way, e.g., via inclusion during design. Behaviour change strategies usually seek to change behaviour for the longer term, i.e., for the life of the project, whereas participatory approaches tend to focus on participation mainly during the project development phase. Illustrative examples for the identified behavior change approaches are presented in Figure 4 and for participatory approaches in Figure 5. Projects tend to aim most often for the strategy of providing information, otherwise, the projects most commonly opt for feedback and monitoring strategies, sometimes combined with incentives or social comparison. There is an interest in the use of gamification also, which often combines feedback with incentives or social comparison.

Figure 6 and Figure 7 show the frequency of occurrence of the main identified engagement strategies. Figure 8 and Figure 9 summarise the distribution of these strategies in the analysed projects.

Participatory approaches used in the projects analysed are diverse. While most approaches are used as part of the project development process (e.g., interviews or surveys to understand user requirements), others may serve as a means of encouraging participation throughout the project’s lifetime, e.g., apps or platforms that facilitate interactive and collaborative virtual communities online. Some strategies could be considered either behaviour change or participatory; e.g., information provision, or social networks, which may combine peer influence or building group identity with a participatory contribution e.g., allowing gathering inputs from participants or facilitating discussions. For simplicity however, we categorise these as one type or the other.

In the following sections, we discuss the engagement strategies used in detail, as well as their outcomes where available. We then summarise our findings and identify potential points for improvement in

Table 1. Summary of behaviour change strategies and points for improvement.

Engagement Strategy	Pros	Cons	Points for Improvement
Information provision	Easy to implement May build interest at the start of a project	Burden falls on organisers Less effective used on its own	Must be used in combination with other approaches to be effective
Goal setting	Shown to be effective in encouraging pro-environmental behaviour in other studies as long as environmental benefits are stressed	Community values impact strength of goals, and these may be heterogeneous Requires facilitation of group goal setting	Combine with participatory approaches to understand group values and set appropriate goals e.g., surveys, interviews, workshops or co-design Combine with targeted rewards or incentives to reduce the conflict with hedonic or gain goals
Group or community identity	May stimulate involvement in community energy projects or pro-environmental behaviours	Requires a pre-existing level of trust	Build community identity and trust by organizing group-based events as a way to foster group identity alongside appealing to community and environmental motives
Peer influence/social comparison	Substantial evidence that peer comparison interventions help reduce residential energy use	Peer interactions and comparisons may not happen of their own accord	Organise platforms or events that facilitate peer interaction Consider linking social comparison to group goals to reinforce social norms
Social modelling	Easy to implement: only one or few people needed to influence the group	If local champion is removed, may cease to be effective	When choosing local champions or similar, ensure that continuity is possible for project lifetime
Feedback and monitoring	Shown to be effective in changing behaviour	Effectiveness varies depending on how feedback is delivered	Use tailored, face-to-face and visualization of feedback
Rewards and incentives	Varying levels of success, more effective than sanctions	Effectiveness is impacted by other contextual factors May be less effective than strategies based on social influence	Combine with participatory methods (e.g., survey, interview, workshop, co-creation) to determine underlying community motivations. Combine with social influence strategies to strengthen impact
Gamification	May provide benefits such as motivation and learning, increased awareness or increasing pro-environmental attitudes	Appears to suffer from declining usage over time More research needed	Combine with incentives to continue app usage Incorporate public appeals e.g., environmental messaging

. We also identify future research avenues and/or which could enhance the knowledge gained from such projects in the future. It should be noted that some projects are still in the early stages and the strategies indicated may be tentatively planned, proposed, or still in progress and subject to change during the project timeframe, e.g., due to unforeseen circumstances such as lockdowns etc.

4.1. Behaviour Change Strategies

In the following sections, the behavior change strategies used in the projects, as presented in Figure 4, are described and discussed in further detail.

4.1.1. Information and Communication

Information provision is commonly used to promote energy-related pro-environmental behaviour, e.g., by educating citizens about energy-related problems or raising awareness about solutions [34]. According to the knowledge-deficit assumption, without relevant information, behaviour change will not occur [37]. It is assumed that new knowledge causes attitudes to change, in turn, affecting behaviour [36]. Interaction and collaboration can greatly facilitate knowledge acquisition and learning [44] and can occur during e.g., information sessions, focus groups or workshops. Social networks or forums can also serve as a means to inform participants about project developments and facilitate further interaction and knowledge exchange.

This strategy is used by a majority (20 out of 28) projects surveyed. Information material ranges from the more conventional leaflets, posters, information sheets, press coverage, videos, blogs, digital platforms, animations, graphic tutorials, comic strips to the more elaborate, technology demonstrations, community showcases (e.g., [45]) or dedicated show rooms to demonstrate technologies.

For example, SENSIBLE project made use of visible demonstrations with full-sized equipment templates to give interested participants a clear idea of their likely spatial intrusiveness. Although some participants withdrew as a result, for the rest it provided enough knowledge about the impact of the project on their homes to ensure their confidence in participating. INTERFLEX (Nice demonstrator) used a show room equipped with virtual and augmented reality technologies, where visitors could experience the project and discover the latest technologies and data platforms of the smart city.

The majority of projects surveyed aimed for the low-hanging fruit of providing information, a strategy shown to have little impact on its own [37]. However, it was usually combined with other strategies.

4.1.2. Goal Setting

Goal setting has shown promise in promoting various pro-environmental behaviours e.g., household energy conversation [34] or energy-saving neighbourhoods [46]. Goal framing theory proposes that human behaviour arises from multiple motivations, and that goals guide or “frame” how people think, feel and act [47]. According to the theory, three types of goal (hedonic, gain, and normative) guide environmental behaviour. Hedonic goals guide people to avoid effort, seek pleasure or excitement. Gain goals guide people to improve resources like money and status. Normative goals guide people to act in a way that they think is ‘appropriate’ based on perceived social norms, e.g., keeping the environment clean. If one goal is strongest in a given situation, goal framing theory posits that it will have a greater influence on preferences and decisions, whereas ‘background’ goals will have an influence on the strength of the stronger goal. Pro-environmental behaviour may therefore be encouraged by strengthening e.g., normative goals [48].

While goal setting is often used in combination with other strategies like feedback and monitoring or gamification, usually targeting the individual, this strategy was explicitly described in just 2 projects, COMPILE and SPARCs, which used collective goal setting as an engagement strategy.

As part of their engagement strategy, COMPILE aims to create an energy community in a co-creative process, the first step of which is to define a collective goal for the energy community [49]. In the SPARCS project, the Leipzig demonstrator aimed to maximize of energy savings at the community level via specific energy savings targets, which were bound to social welfare benefits using the cost savings, to be widely communicated to engaged consumers. This project used community goal-setting in combination with feedback and monitoring on individual energy behaviours via an app [50].

The use of normative goals is considered as an important strategy for encouraging pro-environmental behaviours: evidence suggests that people tend to act upon normative environmental goals, e.g., when conflict is removed between the goals or, when normative goals conflict with hedonic and gain goals, the normative goal is strengthened [48]. Research also suggests, for example, that combining goal setting and tailored feedback are effective in reducing energy use [51]. With only 2 projects using this strategy of goal setting, it would hence be worth exploring additional strategies to strengthen commitments or normative goals among community energy project participants.

4.1.3. Group or Community Identity

Group member’s behaviour, including pro-environmental behaviour can be influenced by perceived group values [52]. Creating a shared identity for the group leads to internalization of group goals for the members, motivating them to behave in line with these goals [20]. This is because group norms can influence personal values. e.g., while an individual with personal biospheric values may have pro-environmental behaviour, the behaviour will become stronger if individuals perceive biospheric values as a group priority. Individuals may also take on pro-environmental group values even when previously they were not so important to them [53]. Communal and environmental motives appear to be more important than financial benefits in motivating people to participate in collective energy projects [11]. Hence, involvement in community energy projects may be encouraged by appealing to communal motives. This may also promote sustainable behaviour among people who may not initially have been pro-environmental.

Group participatory strategies such as workshops, training, focus groups or community events are more likely to build trust and strengthen community identity. Trust is an important factor impacting willingness to participate in community energy [33]. Successful energy communities are more likely when there is interpersonal trust and a strong identification with the community [32].

While building group identity may occur as a side-effect of many strategies involving group activities, only a few projects (COMPILE, REACT, LIGHTNESS) explicitly sought to build group identity using social networks, or through the use of a common platform. E.g., the REACT project an online community is part of the user engagement strategy, with the purpose of building unity and a sense of belonging among the users as well as providing insight with qualitative data, encouraging and assisting the collective learning processes involved in self-management of energy and, creating a support network [54]. CIVIS project also highlighted the importance of including this approach, with user feedback expressing a desire for more features of the app that promote a sense of community.

While group or community identity is not explicitly stated as a behaviour change strategy in most of the projects surveyed, we consider it as a side effect of other strategies such as social networks or platforms. ‘Soft’ factors, such as fostering a sense of the common good, rather than ‘hard’ factors like price and cost reductions, are shown to have more influence on participation in energy projects [55]. Understanding group values is important since motivations for pro-environmental behaviour are tied up in individual identities as well as group identities [54] notes the importance of interaction with other group members as a way to form this group identity. Strategies that aim to build biospheric values in a group may encourage people to behave in line with such values. Alternatively, knowing that a group has stronger hedonic or egoistic values may help design appropriate incentives tailored to the community [56]. In the projects we surveyed, group values remain largely under-investigated, which suggests a risk of key design needs being overlooked. Strategies based on social influence, such as social modelling or peer influence, are shown to be comparably more effective than other strategies in a number of studies [37]. A recent review of behaviour-change interventions across all policy sectors found that for the energy sector, interventions based on social influence (e.g., group norms or peer influence) had more success than those that relied on nudges or incentives [37]. However, it should be noted that the effects of interventions are not consistent across studies and very much depend on context and the behaviour that is being targeted [19]. Since some of these strategies (e.g., group identity) are only employed in a few projects that we surveyed, it would hence be worth exploring these strategies and their outcomes further in future initiatives.

4.1.4. Peer Influence and Social Comparison

Humans tend to follow the behaviour of others [19]. Bandura’s learning theory assumes that people observe and imitate the behaviour of others when it is easy to imitate and understand and when the outcomes are relevant and meaningful to the individual [57]. The behaviour of community members can influence others and stimulate behaviour change [37] and the opinion of other members of a community is shown to be important in influencing the formation of public perceptions about energy projects [53,58]. Information transmitted via interpersonal exchanges within an individuals’ social network [37] is more likely to influence behaviour because it is usually more positively received, understood or recalled compared to educational information or even expert advice [19].

Interventions based on social comparison influence participants’ behaviour though comparing their performance to that of other community members. Social norms—descriptive and injunctive norms have an important influence on behavior [59,60]. Appealing to descriptive social norms, for example can encourage energy-conservation behaviours, e.g., by providing information that compares an individual with the behaviour of other people in the neighbourhood or community [19].

Adding an injunctive norm can support behavioural change [61]. e.g., sending a message that an energy practice is good for the community via house colour codes, or comparing individual energy use to the average, Substantial evidence shows that peer comparison interventions may help to reduce residential energy use [60,61,62].

Seven projects used this strategy, sometimes in combination with gamification (e.g., ECREW, LIGHTNESS). The ORIGIN project, for example, in addition to energy monitoring, also mapped the buildings that participated in the project and colour coded them showing energy consumption in relation to the average, e, using normalised indicators.

In the CIVIS project, the user study revealed that people are willing to share data (publicly or with a selected group of people) about their energy behaviour. Swedish demonstrator participants were interested in knowing the average energy use or building-level energy reduction actions in other cooperatives. Italian test site participants wished to have daily consumption comparisons between participating municipalities [63].

The FLEXCOOP project showed that peer comparison is of great interest in energy cooperatives, and motivates energy efficient behaviour.

4.1.5. Social Modelling

A particular type of peer influence is social modelling. Interventions involve using a ‘model’ person to demonstrate a certain behaviour, in order to influence others to behave in the same way [37]. Carefully choosing key individuals to engage is hence likely an effective strategy for changing the behaviour of a group [64]. “Local champions” for example, are members of the local community with certain competencies who play a role in project endorsing and running a project or supporting participants [65].

This strategy is used by 4 projects (HESTIA, IELECTRIX, ORIGIN, SENSIBLE). For example, the ORIGIN project created local champions who then became trusted promoters of the project in each community. Evidence suggest that without them, the notion of demand response was forgotten by the community and energy practices returned to how they were before the project [66], showing the importance of community-based strategies in achieving sustained changes.

Similarly, using local champions allowed researchers from the University of Nottingham to build on existing rapport with residents enhancing communication, trust and community participation in project SENSIBLE.

4.1.6. Feedback and Monitoring

Feedback involves informing individuals or groups about their behaviour, in comparison to a pre-defined goal or standard, thus hopefully enhancing feelings of self-efficacy or motivation [57]. This strategy is employed by 15 projects and is one of the most popular strategies among projects to encourage behaviour change of participants. Information about local energy production and consumption is accessed via a user interface, found in participants’ premises, in communal areas, or on phones and tablets. Feedback may also be combined with social comparison or carried out in a community context [67]. Feedback interventions are commonly used to promote energy-related behaviour change, may use various mediums (digital, paper, face-to-face, etc.) and diverse strategies such as monetary information, real-time or historical feedback, disaggregation or gamification [68]. Typically indicators and metrics about participants’ collective performance e.g., demand flexibility potential, self-consumption levels, or grid imports-exports are shown (often along with historical data).

In the HESTIA project, part of the Demand-Response setup is to provide feedback to householders in order to make manual or automatically load-shifts [69]. In the ORIGIN project, local renewables and demand forecasts were colour-coded into levels. The colour codes and textual suggestions (similar to social comparison) prompt participants to change their behaviour. In the project WISEGRID, participants may access individual and collective consumption overviews, as well as neighbourhood solar energy production. In times of high solar production, participants are prompted to consume accordingly.

While most projects do not report on the success of feedback and monitoring strategies, recent research on feedback (i.e., providing energy information specific to the household and its behaviour), suggests that it may effectively change behaviour, albeit with varying levels of success. Tailored feedback is more effective overall [37]. There is also some evidence that face-to-face interaction when receiving feedback is the most appreciated by participants and that combinations of visualisation and tailored information are likely to be more effective [51,67].

4.1.7. Rewards and Incentives

Rewards or incentives provide encouragement to reach common societal and community goals [70]. Injunctive norms tend to motivate and constrain behaviour through the promise of social rewards or sanctions. Usually these comprise financial rewards based on participants’ collective achievements and are usually channeled into initiatives that benefit the whole community. Non-monetary incentives include non-exchangeable gifts or vouchers. Rewards or incentives have been shown as more effective than sanctions in encouraging pro-environmental behaviour, however, they tend to have short-term effects [36].

This strategy was used in 8 projects. For instance, in the ISLANDER project, a demand-response app seeks to incentivise individual consumers to follow the most convenient consumption patterns considering actual and forecasted status of the island’s electrical grid. The app also sets incentives to save energy which in turn help to reduce the energy bill [71].

The CITYOPT project designed a community-based application that sends demand response requests to participants during network peaks. Participants gain points for collective load shedding, which may be invested in projects benefiting the local community [72]. Project participants were highly motivated to reduce their consumption in response to demand response requests. Load shedding per participant was shown to be higher than that of the GRID4EU project (Nice demonstrator), which had bigger consumers, using mostly electric heating.

CIVIS project participants reacted positively to the community-level incentive, reinvesting savings from efficient energy use in the community. However, responses varied between test sites with closer communities reporting higher engagement and satisfaction levels [63]. The CITYOPT project showed that small towns and villages seem to be more cohesive social clusters than city neighborhoods and are more motivated by rewards at the local community level [72].

While 8 projects are successfully using rewards and incentives, the impact of non-monetary incentives on household energy behaviours is still unclear and research is needed on whether they are sustained over time, and whether rebound effects are likely to occur [73]. Compared to other strategies, evidence suggests that incentives are not particularly effective [37]. It is argued that they do not impact on attitudes and hence intrinsic motivation. BRIDGE project analysed a large number of EU energy R&I projects and found for instance, that in wealthier areas, monetary incentives are less motivating [25]. Also, cash and non-cash incentives have different effects depending on the type of behaviour [37].

4.1.8. Gamification

Gamification uses games or game elements (e.g., rewards) in non-game settings as a means of educating or changing users’ behaviour patterns [74]. Gamification may combine several interventions together such as information provision, goal-setting, social comparison or trust building [35]. Gamification interventions have been used to promote pro-environmental behaviours in various sectors, including environmental education, consumption awareness and energy efficiency behaviours [38]. Reviews of gamification in energy conservation projects, for example, show that it may provide benefits such as motivation and learning, increased awareness or increasing pro-environmental attitudes [35].

This strategy is used by 4 projects, with some interesting insights. For example the ENERGAWARE project used a serious game to teach users, best practices for saving energy at home. Certain energy saving behaviours were found to improve, however positive behavioural change was not found to last to the final stage of the project [75].

The project CITYOPT demonstrated the benefits of combining gamification with community rewards, showing increased motivation for reduced consumption following demand response requests [72]. ECREW Project makes use of PEAKApp, an app designed as a ‘serious game’ similar to poker aiming to increases player’s awareness of their energy consumption and to save energy and money. Every day, players ‘bet’ on their consumption. Player earn more points with lower bets. All points are lost if daily consumption exceeds the bet. Points earned can be exchanged for gifts in the shop. Comparisons to other households’ electricity consumption are also included in the app [76]. Findings of previous PEAKApp studies showed that the analysis, benchmarking, and gaming functionality overall encouraged optimised electricity consumption through better informed consumption choices. Public appeals, like environmental messaging, positively and strongly impacted households’ consumption decisions, however, some households never used the app at all and usage declined over time [77].

4.2. Participatory Strategies

In the following sections, the participatory strategies used in the projects, as presented in Figure 5, are described and discussed in further detail. The strategies are listed in order of their associated level of participation. Information provision is not included in this section as it has already been covered in Section 4.1 and is generally considered as the lowest level of participation.

4.2.1. Surveys, Interviews and Focus Groups

Surveys, interviews or focus groups are means of gathering in-depth qualitative data on different topics. Surveys do not require interaction, whereas interviews or focus groups require one-to-one or group interactions, before or during the project implementation. Surveys are the least difficult approach to organise, which may explain why they were used by so many of the projects.

Surveys were used by 16 projects to gather different types of information, from technical information to user requirements. Several projects (e.g., HESTIA, CITYOPT, ENERGAWARE, COMPILE, SENSIBLE) used a survey to gather diverse information such as demographics, technical aspects of homes, energy related practices, behaviours, awareness, preferences and motivations in order to inform the project design. Diary surveys over a longer time (e.g., project ISLANDER) can be used to complement smart meters data collection, in order to collect data on how citizens are using energy and design technologies accordingly [78].

Interviews were used by 11 projects and focus groups by 3 projects (e.g., ENERGAWARE, CIVIS, HESTIA, CITYOPT) in order to gain additional feedback on energy practices, perceptions of the project or business models, attitudes, willingness to participate, changes in understanding or awareness, drivers of or engagement in new behaviours. They may also be used to gather user requirements, improve usability or identify barriers or analysed and used to create a community engagement and implementation strategy [79].

CITYOPT carried out contextual interviews in end-users’ households to understand local people’s behaviours and understanding of domestic energy usage, explore people’s needs, priorities and expectations for the software solution. Based on interviews, they created ‘personas’ that increase understandings of user requirements.

“Contextual observation”, an ethnographic technique, was also used by 2 projects. Contextual observation aims to understand people’s behaviour and needs in real-time settings, by observing and recording how they interact with products, services, environments or processes. For example, CITYOPT researchers observed or ‘shadowed’ participants in their working environment and described their daily activities, tools or equipment used, and indoor comfort issues. In HESTIA project people have the opportunity to contextualise their experiences and for the researchers to get a visual and sensory appreciation of their practice [80].

4.2.2. Workshops

Workshops were used by 18 projects for a variety of purposes, from information provision/show casing with informal discussions to training, to structured or moderated discussions with the intent to gather specific outputs. Workshops may be once-off or held repeatedly during the course of the project. In some cases, workshops were part of a larger co-design strategy with users whereas in other projects they are used in a less structured way.

CITYOPT (Nice demonstrator) project for example used one participatory workshop to investigate people’s experiences and attitudes, using tools like board games as a means of investigating drivers, barriers, behaviours and innovative ideas. The EEPOS project used two stakeholder workshops per site to gauge interest in the project, and investigating whether the design of the technology fitted the needs and expectations of the stakeholders and whether they had the required resources, etc. In the COSSMIC project, several user-centered design workshops were used to produce prototypes through co-design processes with users.

4.2.3. Social Networks and Forums

Like gamification, social networks and for a can integrate various participatory interventions and serve a number of functions. Online social networks and purpose-built forums help to promote knowledge exchange and community building, over the longer term. They also host interactive and collaborative virtual communities of practice online. This strategy was used by 6 projects.

In the INTERFLEX (Nice demonstrator) project for example, Twitter was used to engage the prosumer community, and encourage participation. The account provided regular information about the demonstrator, allowed sharing of publications and interaction with the project team. In the CIVIS project, an ICT platform with commenting and question function, supported collaboration and sharing of experiences and showed numbers of participants taking actions. In the EEPOS project, a collaboration tool linked the project ICT platform and social media (e.g., Facebook, LinkedIn and Google+). The tool allows end users to publish their energy saving and comparisons with other users performance. Participants in the CIVIS project sites seemed to enjoy social network and forum features, supporting the notion that not only individual pre-existing motivations (financial, environmental) drive engagement. Household experiences and interactions, supported by ICT, during participation in the project itself may also serve to enhance engagement [63].

4.2.4. Public Events and Activities

Various types of public events or activities are used by projects as a means of engaging participants. These may be more or less participatory and may serve a variety of purposes, such as providing information, building a sense of community or group identity, networking, building trust or collaborating in decision-making. E.g., science or energy festivals (ISLANDER, IELECTRIX, ATELIER), citizen science activities [81], artistic interventions, walking tours, hackathons [81], competitions [45], science classes [82], street markets or short serious games (INTERFLEX—Nice demonstrator).

4.2.5. Co-Creation or Co-Design Approach

Co-creation or co-design approaches are used in 6 projects, to a greater or lesser degree and may employ various participatory strategies together. It can be argued that any project that attempts to include users in the design process in some way is using such an approach, however for the purposes of this study we include projects in which there is a significant effort to include users. In some projects (e.g., CIVIS, HESTIA, ATELIER), the entire project design is built on a philosophy of co-creation or co-design (or related approaches e.g., participatory design, user-centered design) whereas in other projects citizens may be included in a less structured way in one or more of the design stages of initiation, design, implementation, or evaluation.

For example, in the project FLEXCOOP, end user’s requirements were gathered through a participatory process held in the two FLEXCOOP Living Labs. COSSMIC project, via surveys, workshops and visits to users, assessed the user-friendliness of the installed software, the influence of the entire system on lifestyle/active involvement, and other related issues as experienced by the users. Project concepts were developed using co-design workshops and games involving participants. Approaches such as rough prototyping, experience prototype and product box were used, along with communication tools such as tomorrow headlines, storyboards, and service images [83].

HESTIA project uses a Participatory Design (PD) methodology to engage households as collective social units rather than individual consumers, framed by the theoretical considerations of social practice theories [84].

Similarly, the ORIGIN project set up an end-user advisory group to take part in participatory design workshops.

The CIVIS project, use of a co-design approach also revealed the potential for creating unfulfilled expectations, which could negatively impact on participation and engagement. Italian test users were unsatisfied that their inputs were not reflected in the final design [63].

4.3. Summary

Our analysis shows that EU-funded energy projects involving citizens at the collective level are using a variety of behaviour change and participatory strategies in order to engage participants through the project life cycle. In most cases, projects are using a combination of both types, and may use several of each type together, which makes it difficult to measure the effectiveness of individual strategies. In any case, since most of the projects surveyed do not generally offer insights into the longer-term success of their engagement strategies, it is difficult to conclude which strategies are likely to prove most effective and therefore difficult to pinpoint best practices for future projects.

With regard to behaviour change strategies, Table 1 summarises our points for improvement based on our analysis of projects surveyed supported further by the literature.

With regard to participatory strategies, overall, the projects tend towards ‘mid-level’ engagement or ‘consultation’ according to the participation ladder in Figure 3. A majority of the projects consulted or included participants in some way, either via surveys, interviews or workshops, whereas fewer projects took a fully ‘co-creative’ approach. While most of the projects used participatory methods, the degree of involvement of the participants varies greatly and it is not always clear to which extent the outputs of these interventions are used in the projects. In order to avoid reproducing existing differences and power relations, the governance of community or collective energy actions must be carefully considered [26]. BRIDGE (2021) recommended that tailored mechanisms of governance based on democratic decision-making and transparency are needed to ensure the continued engagement of citizens in collective energy projects. Since co-creation is regarded as particularly suitable for promoting decentralized collective energy actions, it may be appropriate to encourage this approach in future projects. For example, the CITYZEN project reported that it was difficult to engage with citizens that had no influence over the way renovations in their houses would be done. They propose that methods giving citizens more voice in the decision-making would be beneficial in future [85], which implies the need for a co-design approach. While some projects report KPIs related to their participatory strategies during the project (e.g., attendance to events) it is difficult to assess the longer-term success of participatory strategies, since long-term engagement with the project or associated tools is not reported in most cases.

4.4. Future Research Avenues

Our analysis reflects a general trend in the literature on collective action energy projects, i.e., that local impacts are not being formally assessed, making it difficult to assess and draw conclusion about the effectiveness of engagement methods used. Furthermore, general good advice for creating better behaviour change interventions include establishing a control group, starting small by pre-testing a smaller group, not overlapping treatments, clearly defining success measures, and monitoring for the longer term [37] is not being followed in many cases.

In general, it appears that while aiming for behaviour change or increased engagement, only some projects base their design methodology on relevant social science theories or take a planned and structured approach to engagement and monitoring impacts. In some cases, engagement or participation seems to be included as an add-on or afterthought, and not well connected to other parts of the project. There is great diversity in the way different projects account for the social dimension and measure social impacts (if at all). This may mean that the desired social impacts are not achieved by such projects in the longer term. The qualitative outcomes of collective energy projects remain understudied for a number of reasons: conceptual clarity on the definition of social impacts of collective energy projects is lacking, the scope of local impacts are unclear, and social impacts rarely have linear cause-effect relationships and hence may go uncaptured [86].

Future research projects should aim to better conceptualise and capture social impacts and hence determine the factors of longer term success of their engagement strategies, while taking account of the influence of cultural or regional effects [86].

As well as this, little research has been carried out for the longer-term impact of pro-environmental behaviour change interventions e.g., incentives [37] and the various factors influencing motivation to participate in collective energy actions are still not yet fully understood [11,19]. Certain behaviour change strategies are only employed in a few projects that we surveyed, (e.g., goal setting, group or community identity) it would hence be worth exploring these strategies and their outcomes further in future initiatives. Future collective action energy research projects could contribute, for instance, to the knowledge on the influence of group values to the success of various behaviour change strategies. Understanding group or community values from the beginning may also result in designs that are more in line with the needs of the communities in question. This is also a key goal of co-creative design approaches in general. Similarly, strategies incorporating community goal setting are also employed in just a few projects.

The questions of inclusion or representation are also important to address, since they have important implications for energy justice outcomes. In terms of participatory interventions, ensuring citizens have adequate influence on decisions that directly affect them can result in a more successful project. The influence of different levels of participation on project success is an understudied research area. In general, future research projects could contribute more to the field of socio-technical research by framing some of their research questions in a social science context from the beginning of the project and attempting to answer outstanding research questions regarding the various engagement methods they wish to use

5. Conclusions and Recommendations

Energy collective action projects are essential to the energy transition. Due to their unique characteristics they can create social bonds and enhance the outcomes of collective energy projects. Engagement methods offer potential for upholding the principles of energy justice and democracy within the implementation of collective energy projects. Sustained engagement of citizens in energy R&I projects is key to their success but remains challenging. We examined the engagement methods used by a selection of collective energy R&I projects and found that a combination of behaviour change and participatory strategies are being used, with the most popular behaviour change strategy (apart from information provision) being feedback and monitoring. We identified potential improvements that could be achieved by modifying or combining different engagement approaches, based on the available project results and latest literature. We found that the majority of projects use a participatory approach, but implementation varies greatly between projects, with few taking a fully co-creative approach, suggesting that there is more scope for citizens having a voice in project decision-making. These findings can help collective energy initiatives to develop engagement strategies that lead to sustained commitment and lasting behavioural change.

For most engagement methods, the project results or literature do not (yet) offer clear conclusions on their effectiveness, which in many cases depends on the context. This means there are significant knowledge gaps still waiting to be filled. R&I projects allow experimentation with different engagement approaches and to identify challenges or best practices, however since the projects we surveyed do not always report coherently on the social impacts of their projects, we miss the opportunity to contribute to the knowledge on these topics. Social impacts have their own unique characteristics and associated challenges, which may not always be fully understood in the context of technical projects. By introducing additional reporting requirements and providing guidance on appropriate methodologies for assessment and reporting in a socio-technical context, future EU-funded calls for proposal can ensure that collective energy R&I projects will be better placed for addressing future challenges of the twin green and digital transition. We therefore recommend that future EU-funded calls for proposal:

• Require more comprehensive results from EU-funded R&I projects in order to determine their effectiveness
• Require longer-term social impact monitoring plans related to broader policy goals
• Clearly specify requirements for reporting on and monitoring social impact so that they are easily comparable between projects
• Encourage proper integration of state of the art social science in project methodology
• Encourage projects to answer social science, as well as technical, research questions related to methods they are using or trialing
• Encourage the use and research of co-creative approaches