For Young and Future Generations? Insights from the Web Profiles of European Climate Pact Ambassadors

Abstract The European Climate Pact provides opportunities for individuals, communities and organisations to declare their commitment to climate action. This study analyses the publicly available web profiles of the European Climate Pact Ambassadors (PAs) as of January 2023. First, it explores the extent to which people who volunteer as PAs demonstrate commitment to young and future generations. Second, it investigates whether PAs who self-identify as young people are more likely than other PAs to justify their mandate by referring to the interests of young and future generations. Third, it examines whether PAs who self-identify as young people are more likely to indicate other young people as the target audience of their activities. The manual coding and quantitative analysis of the PAs’ web profiles revealed that members of older generations as well as parents and grandparents are most likely to rationalise their engagement in the programme by referring to young and future generations. The data also showed that young people do target other young people when they act as PAs, but they are not the only group to do so. When compared to individuals with other professional identities, educators are also more likely to flag young people as their target audience.


I. Introduction
Climate change represents one of the principal threats to young and future generations and, as such, has become a key issue on the regulatory agenda of the European Union (EU).Some law scholars argue that the ascent of climate change has not only modified the substantive content of EU regulatory policies but also established a new paradigm for risk regulation, which, in contrast to standard regulatory responses to conventional risks, 1 "must target systemic change instead of stability, and must favour the virtues of integration and orchestration over those of individualisation and compartmentalisation". 2 Climate change has not only arguably changed the design of the EU's regulatory responses but also its approach to regulatory governance, which has become more open to governance innovations. 3he European Climate Pact arguably represents such an innovative governance approach.It attempts to incorporate the perspectives of different people, communities and organisations in the process of significantly accelerating the reduction of greenhouse gas emissions to achieve "climate neutrality" by 2050 as enshrined in the European Green Deal. 4 Among these perspectives is that of young and future generations, who climate activists have increasingly mentioned as a group for which action needs to be taken.At the same time, young people themselves have started to call for climate action.
The degree to which the European Climate Pact deals with young and future generations is of both academic and practical relevance.It is conceivable that the public is more willing to bear the costs of climate policy5 if these are presented as necessary to protect young and future generations. 6Similarly, climate activism could be conceived as more credible if it is seen to protect the interests of this group.
At the heart of the European Climate Pact are the Pact Ambassadors (PAs)individuals who "inform, inspire and support climate policy and action in their communities and networks". 7Our study focuses on PAs as individuals and draws upon the information they provide on a dedicated website about their personal and/or professional background, their motivation for acting in this capacity and the activities they intend to carry out during their term.We strive hereby to answer the following research questions: • Which PAs refer to young and future generations in their statements about their motivation for engaging with the European Climate Pact? • Which PAs identify young people as the targets of their activities?
Numerous concepts and literatures reflect on the question of how the interests of certain societal groups are represented.For example, in the literature on political representation, the theory of a "politics of presence"8 postulates that a representative's personal background and experiences have an impact on their behaviour. 9This theory suggests that the interests of different groups are best incorporated in policy processes if they are represented by members of their own sociodemographic group.A further perspective is offered by social identity theory, which posits that individuals who share an identity with a social group will align their actions in a way that it is beneficial to that group. 10Recently, Hornung has shown that social identity theory explains climate action by individuals. 11In line with this reasoning, we expect the social identity of individuals to explain whose interests they claim to represent and how they choose the target populations of their actions.
To answer our research questions, the next section provides background information on the ambassadors' programme.Subsequently, the theoretical argument is presented, which is followed by explanations of the methodological approach.Then, we present the empirical findings, before discussing them in the concluding section.

II. Background information on the Pact Ambassadors
The ambassadors' programme has existed since 2020. 12PAs are members of communities, civil society, organisations, education and cultural institutions.Individuals eligible for becoming PAs are "mayors, parliamentarians, policymakers, and other public office holders with a commitment to climate action and ambition", and they "must be a community/organisational leader, or a public servant currently in office". 13ll PAs must make a pledge on behalf of their (informal) group or organisation.The pledges must go beyond existing legislation and either represent ambitious actions that contribute to greatly reducing greenhouse gas emissions and meeting the goals of the Paris Agreement or encompass a wide range of climate-friendly activities. 14The Commission can directly sanction under-performance or non-compliance with the pledges, and deviations from the pledges can potentially affect the individual PA's future engagement with the European Climate Pact.
PAs must also respect the Pact's values, which include science-based actions, transparency, abstaining from greenwashing, an adequate level of ambition that reflects urgency, diversity, inclusion and action tailored to local contexts. 15PAs act in their personal or professional capacity and not on behalf of the European Commission.They also do not receive any organisational or financial support from the European Commission. 16As are expected to lead by example, inspire others to act and connect with others. 17Along these lines, the Commission suggests that PAs organise activities related to environmental and climate action as well as test climate solutions and assess how they can best be replicated and diffused.PAs are expected to promote climate action in their networks, participate in public and private events and organise events and discussions with policymakers.The third aspect of their mandate concerns connecting local climate activists with other relevant networks and stakeholders and facilitating peer-to-peer learning through knowledge exchange.
In return, PAs will benefit from receiving recognition as well as access to networks, resources, toolkits, materials, events and policy experts.The Commission also explicitly states that PAs will have opportunities to make their climate action visible and to exercise advocacy. 1812 J Tosun, "What role for climate pact ambassadors?A policy process perspective" (2022) 21(2) European View 171; J Tosun, J Pollex and L Crumbie, "European Climate Pact citizen volunteers: strategies for deepening engagement and impact" (2023) Policy Design and Practice 10.1080/25741292.2023.2199961. 13European Commission, supra, note 7. 14 ibid. 15Tosun, supra, note 12. 16 Tosun et al, supra, note 12. 17 ibid. 18ibid.

III. Theoretical argument
We postulate that individuals can act in many social roles to affect climate change. 19urthermore, we contend that individuals are aware of these social roles; in other words, that they possess a social identity 20 and express their self-identification when rationalising why they volunteer as PAs.The literature on political representation suggests that those social identities are rooted in individuals' experiences as members of socially constructed groups and thereby motivate political behaviour. 21tudies on social identity, which encompasses social identity theory and its extension to group cohesion, known as social categorisation theory, 22 indicate that group membership enables identification with specific group values, which has arguably been related to some of the polarisation surrounding climate change action. 23Concerning the latter, responsiveness to different framings of communications about climate change policies has been shown to be greater when endorsed by ingroup members. 24Identifying as part of a group can make individuals feel a sense of collective efficacy greater than that at the individual level. 25In other contexts, group leaders (encapsulating some of the key characteristics of their ingroup) may mobilise their group followers by redefining the group's collective identity. 26And existing research suggests that young people tend to be more strongly oriented towards their own generation than future generations. 27n fact, representing and formulating policy demands in the name of future generations is challenging because these cannot speak for themselves.Instead, others must advocate their interests. 28However, in the various social movements for climate action launched in response to Greta Thunberg's school strikes, young people have demanded swift and ambitious climate action in the name of their own generation as well as of future generations. 29Thus, we formulate the following study hypothesis for our first outcome variable, which combines referrals to young and future generations: H1.PAs who self-identify as young people are more likely to refer to young and future generations as a justification for climate action in their web profiles than their counterparts who identify themselves with a different social role.
The second outcome variable of this study is whom PAs consider to be the target population of their ambassadorship and therefore of their communication and mobilisation activities.Policy research has compellingly argued that the choice of target population is important for the effectiveness of public policies.Identifying too many targets can be as problematic as excluding some targets who would contribute considerably to the achievement of policy goals. 30Of course, with the types of climate action PAs carry out, there is no need to formally identify target populations.However, there is value in assessing which PAs explicitly identify young people as the targets of their activities because this is also a way to act in the interest of future generations, as young people represent the future "makers" and "takers" of climate policy.
H2. PAs who self-identify as young people are more likely to refer to young people as targets of their activities than are their counterparts who identify themselves with a different social role.
The empirical analysis focuses on social identities, but we will also test the effect of professional identities to ensure that we capture all identity-related constructs.

IV. Methodological approach
The European Commission publishes each current PA's profile on a dedicated website, 31 which includes the following information: • Type of ambassadorship; that is, whether it is an individual mandate or a mandate on behalf of an organisation (eg European body, government agency or private company) • Name of the PA • Contact information • Mission statement (entitled "Responsibilities" on the website) • Thematic areas (indicated through the selection of standardised keywords) • Country of residence The PA's term usually runs for a year.On its website, the European Commission only presents information on the current PAsno information is available on those who have already completed their term.Consequently, the information provided by the Commission is sensitive to the time of extraction.For this analysis, we use a dataset comprising 876 individual PA profiles downloaded from the European Commission's website on 16 January 2023. 32or analytical purposes, the PA's mission statement is the single most interesting and important piece of information in their web profile.The statements vary in length but do not exceed 100 words.Given the space limitation, PAs must choose carefully what information to present.Research on the self-reported bibliographical notes of politicians has shown that such decisions are made strategically 33 : politicians mention aspects that they want their audiences to know and exclude those that are less important to their audiences or which may not align with their mandate. 30H Ingram and A Schneider, "The Choice of Target Populations" (1991) 23(3) Administration & Society 333. 31 European Commission, "Meet our Ambassadors" (2023) <https://climate-pact.europa.eu/ambassadors/meet-our-ambassadors_en>. 32Ethical approval received from UEA SCI S-REC. 33K Marcinkiewicz and M Tepe, "Positions, Factions and Mandates: Applying Quantitative Text Analysis to Self-Reported Biographical Notes from the Members of the 17th German Bundestag" (2012) 6(2) Methoden -Daten -Analysen 99.
We therefore assume that PAs present only information in the mission statement that is important to their audience, describing what motivates them, why they think they are qualified to act as PAs and which activities they intend to implement during their term.Of course, there can be exceptions, and some PAs may use the mission statements to display other types of information, but overall they present details relevant to their mandate.
We manually coded the mission statements of all 876 PAs to produce the variables of interest for this analysis.Table 1 gives an overview of these and provides example statements for the different coding categories and underpinning decisions.
The first two outcome variables gauge whether PAs motivate their engagement by referring to future or young generations, respectively; the third captures whether PAs refer to young people as the target populations of their activities.All three outcome variables are binary and not mutually exclusive.Given the measurement level of the outcome variables, logistic regression models (with robust standard errors) provide the most apt estimation technique.
The first set of focal explanatory variables refers to the social identity to which individuals assign themselves.We differentiate between members of the younger generation, the older generation, (grand)parents and other.As Table 1 shows, the vast majority of respondents did not indicate any such social role.
In addition to the social identity of PAs, we created a second variable to capture their professional identity, differentiating between academics, people in the private sector, members of civil society (including activists), educators, members of the public sector, students and an "other" category, which comprises individuals who identify themselves with other professions or who indicate more than one profession.The variable also identifies individuals who did not state any professional identity.
Furthermore, we created three control variables.The first one captures whether the PAs represent an organisation, the second one gauges their gender, which we coded based on an assessment of PAs' names, and the third controls for whether they are based in an EU Member State, a country affiliated with the EU or somewhere outside the EU.

V. Empirical findings
Until the end of July 2022, any individual interested in taking climate action, whether residing within or outside the EU, could become a PA. 34The updated eligibility criteria are now stricter and stipulate that only individuals residing in the EU can become a PA. 35In our dataset, PAs based both in and outside the EU are included, as the non-EU PAs were still serving their mandate at the time of the data extraction.
Figure 1 provides an overview of the distribution of 876 PAs, as of January 2023, according to their stated country affiliation.The panel on the left-hand side presents the PAs based in the EU and countries affiliated with the EU, such as Switzerland (a member of the European Economic Area) and the UK (a former EU Member State), and the (potential) accession candidates.The panel on the right-hand side presents PAs based in all other countries.Overall, 720 PAs (82.19%) are based in the EU, 82 (9.36%) in countries affiliated with the EU and the remaining 74 (8.45%) in other countries.Among the EU countries, Italy stands out as the country with the highest number of PAs, followed by Spain and Germany.According to our coding of PAs' names, 47% of them are women and 53% are men, which indicates a fairly balanced representation of these genders.
The descriptive analysis of the mission statements offers some intriguing insights: only about 6% of the total PAs referred to the young generation in their statements and only about 4% to future generations; the share of those who declared young people as their target population is 22% (Fig. 2).
The low share of PAs referring to young and future generations is surprising given the public prominence of youth movements for climate action over the last five years.Most of the 22% of the PAs who referred to young people as their target group refer to young people in the context of education activities, followed by highlighting young people as those whose voices should be heard in climate change governance; however, few give details on how exactly they would attempt to achieve this.Regarding social roles, the PAs coded as "(grand)parents" had the highest proportion of referrals to future generations, followed by "other" and "young generations" (see Fig. 3).Those self-identifying as part of the older generation hardly mentioned future generations.In their mission statements they exclusively referred to the young generation, and in fact of all groups they stand out as those with the highest percentage  share of referrals to the young generation, followed by (grand)parents.In comparison to the identifying groups of "older generation" and "(grand)parent", young generations refer markedly less often to their own (young) generation as a motivation.
Thus, the descriptive analysis reveals that PAs who self-identify as members of the young generation refer less to their own or future generations than do (grand)parents and members of the older generation.However, the picture changes significantly when we inspect the percentage share of referrals to youths as the target group of the PAs' activities, which shows that other young people are mostly the target group of young people (Fig. 3).
Similarly, the professional roles reveal that, with the exception of the PAs who did not state their professional identity, the motivation for volunteering is connected to the young generation rather than future generations.Moreover, for every professional role, the focus on young generations is much more a matter of targeting them rather than mentioning them as a source of motivation.Another noteworthy observation is that educators dominate among the proportion of PAs who refer to young people as the target group of their activities.The share is lowest for PAs who self-identify as members of the private sector.
To supplement these descriptive insights with a multivariate perspective, Table 2 presents logistic regression models.The outcome variables refer to the PAs' stated "Motivations" and "Target group" as identified in our text coding.Because the coding of future and young generations as motivations generated only thirty-eight and fifty-one observations, respectively, we merged both variables into one for the purpose of estimating the logistic regression models.This created a new binary outcome variable, which counted a total of seventy-eight (8.90%) observations in which one of the two motivations or both are present.Therefore, the two outcome variables examined refer to (1) young/future generations as motivations and (2) young generations as targets.
As explanatory variables, the models include the social and professional identities of the PAs as well as three control variables: gender, representation of an organisation and the relationship of a PA's country to the EU.They also control specifically for Italy as the most prevalent country of origin.As an additional robustness test of these choices, we furthermore present in Appendix 1 models that control for country fixed effects.Table 2 presents the odds ratios (ie the exponentiated logit coefficients), which we computed with robust standard errors to account for heterogeneity in the data.The odds ratios tell us how  much higher the odds are of observing positive values (ie a value of 1) for the dependent variable among individuals who self-identify differently.
The first two models (Models 1 and 2) include binary variables for social and professional identities only.These are simplifications of the variables that enter Models 3 and 4, as they indicate whether a PA self-categorized at all.Assessing the effect of those simplified role indicators is already insightful since they show that indicating any social or professional role has notable consequences for whether PAs mention young/future generations as motivations or young generations as a target group.Model 1 reveals that PAs who indicated any social role have significantly greater odds of referring to young and/or future generations than PAs who did not self-identify with a social role.However, indicating any professional role is associated with a markedly lower likelihood of mentioning future/young generations as a motivation (by a factor of 0.36) as compared to PAs who do not state a professional role.Model 2 shows that PAs who indicated their social role in their mission statements have significantly greater odds of flagging young people as the target group of their activities, while the odds ratio for the covariate on professional role fails to reach conventional levels of statistical significance.
Turning to Model 3 with a more refined measurement of social and professional roles, we can see that compared to PAs who did not assign themselves to those roles, members of the older generations are forty-two times more likely to refer to young and/or future generations.We can also observe that PAs self-identifying as (grand)parents and young people are estimated to have fourteen and seven times higher odds, respectively, than do the reference group.This substantiates the descriptive findings presented in Fig. 3 that young people do motivate their ambassadorship by referring to other young people and/ or future generations, but they are clearly not the only ones to do this; in fact, (grand) parents and older generations appear to do so at a higher prevalence.Given the findings of both the descriptive and statistical analyses, we can reject H1.
Turning to Model 4, we can see that, compared to PAs who did not state their professional identity, young people have significantly greater odds of indicating other young people as the target group of their activities.This finding suggests that we can confirm H2.However, it is also worth looking at the findings for the professional identities of the PAs.Model 4 reveals that educators are the PAs showing the greatest odds ratio, suggesting that compared to PAs who did not state their professional identity, educators have almost seven times greater odds of identifying young people as one of their target groups.PAs who classified themselves as students do not have odds ratios that are significantly different from the PAs who did not classify themselves as such.
Turning to the control variables, in Model 2 the odds ratio for gender is positive and significant, indicating that women have greater odds of mentioning young people as their target groups.Yet this finding ought to be interpreted with caution given that the gender variable was coded based on the assessment of names rather than self-declared.The models did not produce significant effects for the variable differentiating between individuals who are PAs in their own right and those who represent organisations.There are some differences between PAs based outside the EU and those based in the EU or in a country affiliated with it.Of the EU-related effects, it is worth highlighting that PAs based in the EU have lower odds of referring to young people as their target group than PAs based outside the EU.
Overall, the findings suggest that the self-identification of a PA can explain whether this individual motivates their engagement by referring to young and future generations as well as whether young people are among the target groups of the activities to be carried out during the Climate Pact ambassadorship.However, it is not only the social role that matters, as postulated above, but also the professional role of the PAs.In combination, these two roles give a reasonably good sense of who represents young and future generations and who regards them as the main group to interact with.
The empirical analysis revealed that young people do advocate the interests of the young and future generations.However, they are not the PAs with the greatest odd ratios of motivating their mandate by referring to these groups.As we show, parents and grandparents as well as members of the older generations tend to have higher odds of referring to young and future generations and explain why they are volunteering for climate action.In addition, we could show that young people are more likely to target other young people than PAs who do not indicate a social role in their mission statement.However, the analysis also shows that educators typically express their interest in reaching young people with their climate actions.

VI. Conclusion
The perceived needs of young and future generations are important for current policymaking.The decisions we make today determine what world young people and future generations will inherit, including societal challenges as well as democratic institutions and policy legacies. 36In this study, we explored how relevant young and future generations are to PAs and their voluntary commitment to climate action.We focused on PAs becauseby taking on the PA rolethese are individuals who are particularly aware of the need for urgent and adequate climate action, and their engagement could contribute to overcoming the "governance trap" in climate policy. 37However, even among the members of this group, the awareness of young and future generations seemed to be limited or at least not expressed explicitly.

Figure 1 .
Figure 1.Number of European Climate Pact Ambassadors (N = 876) as of 16 January 2023 by (a) European Union (EU) country members and associates (left) and (b) non-EU members and associates (right).NA = not applicable.

Figure 2 .
Figure 2. Ratio of European Climate Pact Ambassadors who indicated in their statements that they are motivated by young or future generations, and the share of Pact Ambassadors who declared young people as the targets of their activities.

Figure 3 .
Figure 3. Motivations and target groups of the European Climate Pact Ambassadors (the outcome variables) by the social (left) and professional roles (right) in which they presented themselves in their mission statements.

Table 1 .
Overview of the variables and their coding.

Table 2 .
Logit models explaining future/young generations as motivations or targets.

Table 2 .
(Continued )Exponentiated coefficients (odds ratios; OR); robust standard errors (SE) in parentheses; five observations had to be dropped in the estimation of Model 4 because the social role "older generation" predicted perfectly the outcome "0" of the dependent variable (see Fig.3).