EU executives on social media: Assessing the potential and trajectories of legitimation via public communication on Twitter

Abstract

Social media offers opportunities for European Union (EU) executives to improve their precarious legitimacy. Their broad, unmediated, instant, and cheap reach allows EU executives to explain to citizens what they are doing in their own words and to learn from citizens what they want. Yet, while the potential of social media is well-known, we know very little about how EU executives use them. This study addresses this gap by analysing how EU executives legitimise themselves through strategic public communication on Twitter. Through a manual and automated content analysis of a sample of 73,413 tweets, we show how EU executives engage in ‘opinionated communication’. Appeals to output legitimacy are the single most prominent form of self-legitimation, but they are often combined with opinions, calls for input, and/or emphasis on activities. Our findings challenge the common perception that EU executives are risk-averse communicators who focus exclusively on factual output-oriented communication.

Introduction

The legitimacy of the European Union (EU) is precarious. With the turn from a ‘permissive consensus’ to a ‘constraining dissensus’ in the 1990s, public opinion has become a central factor for the functioning and development of the EU (Hooghe and Marks, 2009; see also Hodson and Puetter, 2019). This has become more evident with increased contestation of the EU in response to a series of fundamental challenges, such as the euro and the migration crises. Many citizens are simply not convinced about the worth of EU membership or the legitimacy of its supranational institutions. EU supranational executive actors are often the direct targets of challenges to EU legitimacy (Silva et al., 2021) but have traditionally been left with few alternatives in responding to such contestation. The landscape of the traditional media in Europe has been fractured by national languages and borders, hindering a genuinely European public sphere. Communication on EU affairs has happened mainly through national actors in national media (Koopmans and Statham, 2010). This constitutes a crucial hindrance for the messages the EU supranational actors try to communicate (Bijsmans and Altides, 2007), making it difficult for them to communicate to a European constituency in a way that might strengthen their legitimacy.

Social media might offer a precious opportunity for EU executive actors to overcome such hurdles. In contrast to traditional media, social media allows EU executive actors to broadcast their messages unmediated by national politicians and journalists (Wallace, 2018), with a cross-border reach (Bossetta et al., 2017). Importantly, many of those EU executives would want to reach use social media platforms, with Twitter (now X) having a particularly important position for political discourses (Jungherr, 2016). The communication of supranational actors on Twitter might also carry on beyond the platform itself as, for instance, journalists use Twitter as a source of news (Cage et al., 2020; Oschatz et al., 2021). We can assume that EU executives are aware of the potential social media provides to boost their legitimacy as existing research demonstrate that social media have become an integral part of the broader EU communication strategy (Ecker-Ehrhardt, 2020; Krzyżanowski, 2020; Özdemir and Rauh, 2022).

Yet, we know little about how EU executives communicate on these platforms. While there are several studies focusing on the nexus of EU actors and communication on social media platforms, the majority of these studies have either been theoretical in nature (e.g., Barisione and Michailidou, 2017; Krzyżanowski, 2020; Zaiotti, 2020) or focused on electorally accountable actors such as members of the European Parliament (European Parliament. Directorate General (DG) for Parliamentary Research Services, 2021; Fazekas et al., 2021; Haman et al., 2023; Haßler et al., 2021; Nulty et al., 2016; Umit, 2017). Meanwhile, there has so far been little research on how the executive branch of the EU communicates on social media. We thus still lack knowledge of how EU executives’ communications on social media can contribute to the legitimation of the EU.

This article contributes to this gap in how EU executives use the opportunity provided by social media by asking how EU executives appeal to the legitimacy beliefs of their audience via public communication on Twitter: Who do EU executives publicise, and what is the content of their tweets? In this article, we study the strategic public communication of 113 EU executives active on Twitter between 1 December 2019 and 31 July 2020 in order to shed light on these questions. Our dataset consists of 73,413 tweets spanning these eight months. By combining insights from the literature on public communication and the legitimacy of international organisations, we build a novel analytical framework to analyse which sources of legitimacy EU executives appeal to.

Based on the existing literature and official EU communication strategy documents, we formulate three conjectures. First, we expect that the EU makes ample use of the opportunity provided by Twitter, attempting to exploit its potential to reach its international constituency without the intermediation of national politicians and journalists. Second, we anticipate EU executives to appeal to a variety of legitimacy sources, including identity, input, throughput and output bases of legitimacy. Thirdly, appeals to output legitimacy are likely to dominate, as EU executives attempt to ‘play it safe’ or not to stir unnecessary controversy in a highly politicised environment. While the evidence stemming from our systematic and rigorous empirical mapping of EU executive communication on Twitter is in line with the first two expectations, it does not confirm the third. Against our expectations, our analysis reveals how EU executives predominantly engage in what we call ‘opinionated communication’: A communication style that combines appeals to output and other forms of legitimacy with clearly voiced opinions and often a personal touch. Perhaps EU executives are less risk averse and more interested in achieving engagement from the public than we so far believed. This is all the more interesting given our additional finding that the object of legitimacy most frequently presented in these tweets is ‘the EU’ as a whole. This implies that possible backlash against this engaging ‘opinionated communication’ could target the entire EU, not just the individual executive, agency or department sending out the tweet. In other words, the communication pattern we document is not without risks to EU legitimacy.

In the following, we first situate our study within the larger debates about the EU's ‘legitimacy crisis’. We then discuss the role EU executives have in facilitating the legitimacy of the EU and how social media, and Twitter in particular, might be a promising venue faced with the so-called ‘communication deficit’ of the EU. This is followed by our framework for analysing how EU executives may appeal to different legitimacy beliefs using strategic public communication, first, by discussing the different sources of legitimacy EU executives might draw on, and secondly, by presenting a framework for how they might do this through the use of strategic public communication. The next part presents our research design, data and method. In the final part, we map the diversity and frequency of the content of the EU executives’ tweets and discuss the potential implications of our findings for the cuing of EU legitimacy beliefs.

The EU's ‘legitimacy crisis’: The role of EU executives and social media

It has been widely claimed that the EU suffers from a ‘democratic deficit’. This argument is based on both normative considerations (e.g., Follesdal and Hix, 2006; Kohler-Koch and Rittberger, 2007) and observations of how citizens’ expressions of discontentment with the trajectory of European integration have become more pronounced, particularly through referendums. On the basis of this, it is claimed that the EU is now facing a ‘crisis of legitimacy’ (e.g., Holzhacker, 2007; Longo and Murray, 2011), a claim that has been further exacerbated by the politicisation of the EU with the euro- and migration crises, and later also the UK's decision to leave the union (e.g., Hobolt, 2018; Schmidt, 2020).

EU executives have been a focal point within these larger debates about the EU's legitimacy. We understand ‘EU executives’ as the actors who are mainly responsible for implementing the policies to follow political goals set by the European Parliament and the Council of the European Union. This includes institutions such as the European Commission, its DGs and agencies, as well as individual actors sorting under these institutions. The EU treaties set out that these institutions and individuals have a mandate to represent the common interest of the EU through designing and implementing policies, but with limited mechanisms for democratic representation. Because of this, the contestation of these actors has been central within the increased politicisation of the EU. A study by Silva et al. (2021) finds that heightened politicisation of European integration in the period 2002–2017 is associated with a stronger, predominantly negative, presence in the media of supranational actors, such as the European Commission and the European Central Bank. EU executives – their technocratic way of operating and lack of democratic accountability to voters in particular – are challenges to the legitimacy of the EU.

However, EU executives are also central players within processes of legitimation. In response to public contestation, they engage in acts of discursive self-legitimation. A growing body of literature argues that the aim of generating and promoting generalised support has become increasingly important for actors representing international institutions, particularly in response to heightened contestation, and that this is reflected in their public communication (see for instance Dingwerth et al., 2019; Ecker-Ehrhardt, 2018; Gronau and Schmidtke, 2016; Tallberg and Zürn, 2019). This is also the case for the EU, with a particular focus on the European Commission (Aagaard, 2023; Bijsmans and Altides, 2007; Meyer, 1999). Although communication and information policy has been on the EU executives’ agenda since the start of the European integration process, it has intensified over time (Nesti, 2010; see also Altides, 2009; Gramberger, 1997). By now, reproducing and maintaining legitimacy has become a sustained strategic goal for the executives rather than mere sporadic crisis management responses. For instance, the DG for Communication (DG COMM, 2020: 7)'s most recent strategic communication plan, covering the period 2020–2024, explicitly points to the importance of securing legitimacy beliefs.

Yet, it remains a question whether EU executives are efficiently capable of fulfilling the aims of self-legitimation. A crucial aspect to the debates about the EU's legitimacy crisis and the legitimation efforts of the Commission more particularly is the idea of a ‘communication deficit’ (Meyer, 1999). It is argued that a main obstacle to the legitimacy of the EU is how the EU lacks the ability to communicate efficiently to a broader public (Ward, 2010: 124–125). This communication deficit has largely been ascribed to two interwoven factors: media dynamics and the EU as a political system.

Firstly, access to communication channels in order for information to reach the intended audience is a challenge for EU actors. Public spheres mostly remain organised along national borders. These national public spheres are guarded by gatekeepers, in particular national journalists and politicians who often block or distort EU messages to citizens (Koopmans and Statham, 2010; Risse, 2014; Trenz, 2004; Walter, 2017). The attention paid to European affairs in national media has historically been moderate and most often linked to specific events (De Vreese, 2001; Kantner, 2004). Furthermore, the coverage of the EU in the media is also often negative (De Vreese et al., 2006). The challenge in getting their messages through is highlighted in a study by Bijsmans and Altides (2007), showing how there is a considerable difference between the messages the Commission sends out and what is communicated in the media.

Secondly, it is argued that the technical and complex nature of the political issues the EU deals with make it difficult for EU actors to communicate in a clear and comprehensible manner (Meyer, 2009), and that dynamics within EU institutions themselves tend to contribute to making messages ‘cryptic’ (Bijsmans, 2014: 187). This argument is further underpinned by empirical research showing how the communication of the EU commission is often very technical and inaccessible (Rauh, 2022; Rauh et al., 2019). Consequently, their messages tend to fail to capture the attention of citizens and fall short of satisfying necessary conditions for legitimacy management, such as transparency and publicity (Curtin and Meijer, 2006; Hüller, 2007).

Social media offers the potential to overcome some of the causes of this communication deficit. Although social media encourages short and simple messages and allows for multimodal communication, this seems not to immediately have resolved the Commission's challenges of communicating more comprehensible messages (Özdemir and Rauh, 2022). However, social media might offer a way to bypass the gatekeeping function of traditional media, thereby partly overcoming the ‘communication deficit’: These platforms permit EU executives to determine which messages to inject into this ‘digital public sphere’ and deliver them unaltered (Barisione and Michailidou, 2017), and messages on social media platforms can easily traverse national boundaries, thus offering an extended reach for the communicator (Bossetta et al., 2017). Existing studies indicate that social media, and Twitter in particular, show potential for a more transnational European public sphere (Hänska and Bauchowitz, 2019). Research on international organisations more generally shows how social media have become important channels for them in communicating with the general public (Bjola and Zaiotti, 2020; Ecker-Ehrhardt, 2024). Yet, we still know little about what messages EU executives send out on social media such as Twitter.

Appealing to legitimacy beliefs through strategic public communication

Public relations as part of self-legitimation efforts

Most empirical legitimacy studies, including those that study how political authorities discursively legitimate themselves, build on the theoretical foundations of Easton's system-theoretical approach (e.g., Hurrelmann, 2019). Easton distinguishes between two forms of support for a political system. On the one hand, there is specific support shown toward individual policies, decisions, and actions. Diffuse support, on the other hand, is a ‘reservoir of favourable attitudes or goodwill that helps members to accept or tolerate outputs to which they are opposed or the effects of which they see as damaging to their wants’ (Easton, 1975: 273). From this perspective, empirical legitimacy, that is, citizens’ belief that a political system is legitimate, is a form of diffuse support where people, who are bound by the decisions of a political system, accept the authority of the political system and follow its rules (Gilley, 2009; Hurrelmann et al., 2007).

Earlier literature has made a distinction between legitimation strategies as relating to diffuse support and public relations as relating to specific support when it comes to discursive legitimation (Gronau and Schmidtke, 2016). We hold that conducting public relations is also an important part of the self-legitimation efforts of international organisations for two reasons. Firstly, the line distinguishing diffuse and specific support is rather thin when it comes to the EU. Because EU executives lack direct democratic accountability to EU citizens, discontent with policies they devise and enact can easily result in discontent with the whole polity of the EU (De Wilde and Trenz, 2012). Therefore, public relations efforts revolving around specific policies and acts can serve to legitimate the EU as a polity.

Secondly, public relations are by definition endeavours of legitimation. Public relations are a strategic management function (Dozier et al., 1995) that aim to ‘establish and maintain a mutually beneficial relationship between an organisation and the public on whom its success or failure depends’ (Broom et al., 2013: 26). Here, the success of a political authority can be defined as the exercise of authority without resorting to coercion or guile (Beetham, 1991). In principle, all political authorities try to foster beliefs that they are rightful, proper, and worthy of obedience (Gerschewski, 2018; Weber, 1978). If unfavourable political outputs can degrade the diffuse support and the successful exercise of authority entails acquiring compliance without deceit or coercion, then public relations operations are a crucial mode of self-legitimation for political authorities.

Public relations efforts can take many forms, from discursive acts to physical events. Strategic public communication is one possible activity for maintaining good relations with a given public (Bäckstrand and Söderbaum, 2018). Strategic public communication is the purposeful use of communication by an organisation to fulfil its mission (Hallahan et al., 2007: 3). In other words, it is a set of deliberate messages for a designated public and delivered via the most appropriate media channels at an appropriate time to achieve a pre-determined effect (Plowman and Wilson, 2018). In this sense, “strategic” serves to indicate that the act is evaluated in terms of contributions to the goals of the organisation (Hallahan et al., 2007: 11). Concomitantly, strategic public communication through various channels, such as social media platforms, can be taken as acts of legitimation.

EU executives and sources of legitimacy

Appealing to legitimacy beliefs with strategic public communication is a rather complex task for EU executive institutions and actors. The legitimation literature identifies four variants of legitimacy sources, which are also reflected in the more overarching debates on EU legitimacy.

The first is input as a source of legitimacy, where the political authority is believed to be legitimate to the extent that it is responsive to the concerns of its constituents (Bellamy and Kröger, 2013; Follesdal and Hix, 2006). Input legitimacy has been central to the aforementioned debates about the EU's democratic deficit, with scholars arguing that EU executives should improve the channels for input and participation by those who are bound by their decisions (Bellamy and Weale, 2015; Borrás et al., 2007; Follesdal and Hix, 2006).

Arguing against the problem of a ‘democratic deficit’, others have focused on a second source of legitimacy, namely output. If legitimacy is based on output, a political organisation is legitimate to the extent that its policies effectively and efficiently contribute to the common good (Scharpf, 1999). Since EU executives possess substantial regulatory capacity, it is argued that the EU and its executives should legitimate themselves based on the efficiency of their output and capacity to contribute to the common good (Hix, 2008; Majone, 1997, 1998; Scharpf, 1999).

The third source, throughput, is placed between output and input in the political system. Throughput legitimacy of the political authority is based on the quality of the governance procedures, such as the efficacy of the policy-making process, accountability and transparency vis-à-vis the relevant stakeholders, and openness and inclusiveness with regard to civil society (Schmidt, 2020). System-wise, bad throughput phase consists of oppressive, incompetent, corrupt or biased governance practices that isolates relevant stakeholders from governance and provides limited or no transparency on the procedures and responsible (Schmidt, 2013). Such governance practices would naturally undermine the beliefs in the legitimacy of a polity. It is argued that EU executives can potentially improve their throughput legitimacy by making the decision-making phase more transparent and open to stakeholders, thereby providing insight into how policies are made (Chatzopoulou, 2015; Schmidt, 2013).

Recent debates on sources of legitimacy have also included a fourth source – identity. In the case of identity legitimacy, a political authority enjoys legitimacy based on what the organisation is, the values it stands for, and how far the essence of the organisation is right and proper (Barker, 2001, 2007; Dingwerth et al., 2019; Suchman, 1995). Identity as a source of legitimacy covers several possible sub-sources such as a mission to uphold liberal norms or to conduct efficient technocratic governance.

In light of this literature review, firstly, we expect EU executives to engage in active strategic communication via Twitter as they have done so in many other communication platforms. Their precarious legitimacy is well known, to themselves as well, and their realisation that communication is important speaks from their investment in professional communication, most notably in the form of DG COMM.

Secondly, since the EU cannot afford to lose the faith of European citizens in the long run, we also expect EU executives to communicate in order to appeal to all above-mentioned forms of legitimacy sources to some extent. The official communication strategy of the European Commission clearly highlights the various bases of legitimacy – identity, input, throughput and output – possibly available to the EU, demonstrating they are well aware of the importance and possibilities of communication in this regard (DG COMM, 2020).

Thirdly, even though EU executives can appeal to these sources via strategic public communication, the scholarly literature has also pointed to another aspect that has to be taken into consideration. Self-legitimation efforts may contribute to the politicisation of European integration (Andrione-Moylan et al., 2024) which could backfire on EU executive institutions. They are therefore frequently seen as agents of depoliticisation, trying to present their actions and policies as in the general interest, a scientific or legal imperative, or otherwise ‘alternativlos’ (Bressanelli et al., 2020). Recent research on how EU executives navigate a politicised environment further underline the often technocratic or neutral communicative attempts they make, attempting to depoliticise EU governance (cf. Bressanelli et al., 2020; Pansardi and Tortola, 2022; Rauh et al., 2019). Thus, we expect EU executives to predominantly rely on appeals to output legitimacy, which has regularly been ascribed to be the main or most reliable basis of legitimacy, both by academic observers and by key EU officials themselves. The question then remains how they do this via public communication in practice and how we as analysts can approach this empirically. We now turn towards this operationalisation, drawing on public relations literature to equip our subsequent empirical analysis.

Cueing legitimacy beliefs via public communication

How can strategic public communication appeal to these sources of legitimacy beliefs? Recent experimental studies show that public communication of the elites often works through the mechanism of cueing (Dellmuth and Tallberg, 2023). As a communication effect, cueing refers to the phenomenon where the audience of the communication rely on elite's communication to form their opinions (Darmofal, 2005). This implies that public communication from EU executives can cue different legitimacy beliefs based on the content of their communication.

Based on the content typologies established in public relations literature (Dozier et al., 1995; Glozer et al., 2019; Grunig, 1992; Mergel, 2013; Taylor and Kent, 2014; Waters and Williams, 2011), Figure 1 illustrates how different content of messages can potentially appeal to various sources of legitimacy beliefs. It should be noted that, even though the illustration seems deterministic, it only illustrates potential cueing effects of message content. The likelihood that these connections occur is beyond the scope of this paper.

Figure 1.

Potential trajectories of legitimation in public communication.

Legitimacy beliefs based on identity can be maintained by messages that inform the audience about the identity and mandate of the political authority. Such messages can potentially educate the audience about the raison d'être, the purpose and the foundational norms, of the authority and thus alter the legitimacy beliefs of the audience by creating the sense that the organisation's identity is desirable and proper. For example, if EU executives in a tweet claim that they are the servants of the people and that they enact the will of Europeans, alluding to liberal-democratic norms of representation, this could potentially inform and cue the audience about executives’ identity.

Output-based legitimacy beliefs can be fostered through communication where the content of the message focuses on the political outputs and activities of the authority. By informing people about the successes and achievements of the political organisation, political authorities can encourage the belief on the part of the public that their political outputs effectively and efficiently contribute to the common good of the audience. Similarly, updating the audience about day-to-day activities and agenda, the executives could cue the impression that they are hard at work, thus contributing to the common good.

Executives can appeal to throughput-based legitimacy beliefs with their strategic public communication in two particular ways. Throughput-based legitimacy essentially rests on two beliefs: First, the audience of the authority should perceive the governing processes of the authority as transparent; and second, the target audience of the authority should perceive that the governance process involves and includes all relevant societal groups (i.e., stakeholders) (Schmidt, 2013). Thus, messages that report on the responsible individuals and authorities in political decisions and outcomes can encourage beliefs about transparency among the audience. Here, responsible actors are understood in terms of those who have a causal role in the design and actualisation of political outcomes (Hobolt and Tilley, 2014: 12), and transparency is understood as reporting these roles in communication. To illustrate, this could be done so by tweeting about the executive actors who spearheaded a policy, or an institution that tackled the design of it. In terms of appealing to the beliefs that relevant societal actors are involved in political processes, executives may again rely on messages acknowledging the role of and reporting on the participation of relevant societal actors in political processes. For example, tagging relevant societal actor in tweets and informing the audience about their participation in the governance process. This might also happen more indirectly, through messages that invite relevant stakeholders to engage in the process and thereby signal that they are included.

Finally, input-based legitimacy beliefs may be fostered by two types of message content. Firstly, by asking for input and feedback on the political goals and outputs from the broader audience, a political authority can establish or enhance positive beliefs about its capacity to represent the interests of its constituency in governance (Bellamy, 2010; Bellamy and Weale, 2015). In practice, this would create another, yet unbinding, input taking channel in addition to more formal mechanisms. Secondly, messages that are replies to the audience could potentially result in positive outcomes such as trust, mutuality, and empathy between the political authority and its constituency (Soon and Soh, 2014; Taylor and Kent, 2014) as the political authority enters into a dialog with its constituency on the basis of equality to discuss and set the political agenda (Taylor and Kent, 2014). Executive individuals and institutions replying to comments and tweets of other users, on some level, would become an informal deliberation process. These outcomes can, therefore, feed into the impression among community members that the political authority has the capacity and willingness to represent the interests of the community.

Data and method

Our sample of executives aims to capture the presence of the EU executives on Twitter as widely as possible. To this end, we have identified 113 verified Twitter accounts that belong to the supranational executive branch of the EU. We scraped Twitter handles using the official webpages, and then manually validated these accounts (the complete list of handles is presented in the Online Appendix). At the time of collection, these accounts constituted the full population of EU executive leadership and institutions present on Twitter. These accounts include commissioners, director generals, deputy-director generals, institutions, and agencies.

After selecting the appropriate accounts, we streamed tweets from these accounts using R, Rtweet (Kearney, 2019), and Twitter API v1 between December 1, 2019, and July 30, 2020. Although this period also includes the onset of the coronavirus disease 2019 (COVID-19) pandemic, we reason that these eight months of daily communication by EU executives provide a set of observations sufficiently representative of their total communication, since crisis has become part and parcel of EU politics (Zeitlin et al., 2019). We return to this in our analysis. The total sample amounts to 91,420 tweets. We then selected a proportional stratified random sample of 4427 tweets following probability theory (Krippendorff, 2018) to serve as training, test, and validation datasets. After the manual coding of this sample, we trained a set of machine learning algorithms to code the whole dataset. After cleaning the duplicates and automated coding mistakes, our analysis dataset comprises 73,413 tweets. Details of the manual coding procedure, intercoder reliability scores, as well as machine learning training procedures, and model performances, are presented in the Online appendix.

We operationalise appeals to different sources of legitimacy beliefs via contents of strategic public communication with Subject and Object of publicity. Definitions, example tweets and the codebook for these variables are provided in the Online appendix.

Appeals to identity-based legitimacy beliefs are operationalised with the category Identity and Mandate, covering content referring to the mandate and foundational beliefs. Messages that make a statement such as ‘this is what the European commission stand for’ are coded under this category.

Appeals to output-based legitimacy beliefs are operationalised with the categories Output and Activity. Messages concerning substantial political outputs such as pharmaceutical strategy in the example in the Online appendix are coded under Output. Activity captures messages from EU executives reporting on more mundane activities such as meetings, workshops, and publishing press releases.

Appeals to throughput legitimacy through reporting responsible agents in governance processes are operationalised via the Subject of publicity variable. The Subject of publicity variable contains four main categories: Self, Other, Compound and None. The categories of the Subject of publicity variable are designed to capture “the agent” presented in the tweet in relation to the tweet's author. So, tweets that present the author as the agent are categorised as Self whereas those that present other actors alongside the author are categorised as Compound. Specifically, the categories Other and Compound are designed to capture the transparency principle of throughput legitimacy as the messages with these categories would report on the responsible actors behind the policy decisions. The other principle of throughput legitimacy, i.e., stakeholder involvement in governance, is mainly captured by the Other and Compound categories. Since the two principles of throughput legitimacy overlap to some extent in our Compound category, we also recorded the twitter handle of mentioned actors for the categories of Other and Compound. However, further automated classification of these mentioned actors showed unreliable performance. Thus, the operationalisation of appeals to throughput legitimacy is limited.

The reason for this operationalisation, rather than a direct category capturing such appeals, is that it proved challenging to distinguish such messages about consultations with relevant stakeholders from other forms of messages seeking input, and so on. This way we capture a core part of throughput legitimacy.

Finally, we operationalise appeals to input-based beliefs in two ways. The first is Input-seeking messages, where tweets ask the Twitter audience to express their opinions on a matter or a policy output either on-site (i.e., Twitter) or by providing a direction to other platforms to do so. The second way we operationalise appeals to input-based legitimacy beliefs is the Reply category. Tweets that reply to other users are assigned this label, but we do not explicitly code for the Reply category as Twitter API already provides this information. An illustration of the coding of an example tweet is provided in the Online appendix.

Results

Appealing to through-put legitimacy beliefs: Responsibility attributions and actor involvement

We start our analysis with a focus on the variable Subject of publicity as a potential cue for throughput legitimacy. As discussed earlier, strategic public communication can potentially cue positive throughput-based legitimacy beliefs in two ways: First, by reporting political actors responsible for political outputs, thus creating transparency in the governance process. Secondly, by reporting on relevant societal actors as active participants in political processes and for outputs, so the impression that relevant stakeholders are involved in the allocation of resources is established and reinforced. Categories coded under Subject of publicity, such as Self, Other and Compound, give us insight into both responsibility attributions relative to EU executives and the involvement of other actors as part of appeals to throughput-based legitimacy beliefs.

Almost all tweets explicitly publicise some actors, indicating either responsibility or involvement. We interpret this as an indication that EU executives do appeal to throughput legitimacy beliefs on Twitter. There is an overwhelming tendency for EU executives to publicise other actors with their communication on Twitter, either directly through their own tweets or by retweeting others. Regardless of their specific role in the executive branch, almost 90% of the tweets from each group of executives mention another actor. Meanwhile, setting aside the Other category, we also see a tendency of individuals, such as commissioners and presidents, to present themselves as responsible actors in the tweets, either as Self or Compound (see the Online appendix for details).

Which actors are publicised alongside the EU executives themselves? While the categories of Other and Compound signal that other actors are involved in governance processes, thereby indicating the second form of appeal to throughput legitimacy beliefs, this cannot be immediately derived from the above results based on how we coded the data. To follow up on this, we investigate the most frequently mentioned other actors in EU executives’ Twitter communication by utilising the manually coded data set, as automated analysis failed to capture this sufficiently.

The other actor EU executives most frequently referred to is, in fact, the EU or the commission rather than stakeholders and societal actors. Figure 2 demonstrates that, by and large, the executives present the EU as a unitary actor: For all categories of EU executives, the ‘EU’ or the ‘EU Commission’ is the most publicised actor. Taken together with the findings discussed above, this is indicative of an obfuscation of the responsibilities of individual actors. Although we find that the EU executives to a considerable extent indicate the agency of actors, and thereby might appeal to throughput legitimacy, such communication is more likely to influence legitimacy beliefs about the EU as a whole as opposed to the legitimacy of individual EU officials and institutions.

Figure 2.

Most frequently mentioned ‘other actors’ by the executives.

In other words, discontent among the audience for a particular political output or decision would then be directly linked to the EU as a polity rather than to individual actors who made the decision, thus potentially fostering Euroscepticism (Mair, 2007).

Appealing to legitimacy beliefs

We now turn our attention to how the EU executives’ tweets have the potential to cue other sources of legitimacy beliefs than throughput. Figure 3 shows that at the aggregate level, EU executives’ communication on Twitter primarily revolves around appeals to output-based legitimacy by reporting on outputs and political activities. About 50% of the tweets contain at least one mention of political outputs, and an equal number of tweets refer to the daily activities of the executives. However, the executives’ public communication is not without potential to cue other legitimacy beliefs. About 40% of the tweets include a message informing the audience about the raison d'etre of the executives, thus potentially fostering audiences’ belief in the executives’ legitimacy based on their identity. Finally, about 30% of executives’ tweets invite their Twitter audience to provide feedback or express their opinion on the political outputs and goals, whereas a little less than 10% of executives’ tweets in this period actually engaged with the audience, as indicated by the number of replies.

Figure 3.

Share of message contents for overall European Union (EU) executive communication.

As our study covers the onset of the COVID-19 pandemic in Europe, we also examine how this might have influenced the messages of EU executives. Müller and Fraussen (2023: 2) have argued that the pandemic was a cross-sectoral crisis, during which it was particularly important for EU agencies to ‘represent their organisations’ capabilities and knowledge in a good light’. To investigate the possible impact of the crisis on EU executives’ communication, we disaggregated our findings on the different objects of publicity by month.

This disaggregation shows that prior to the pandemic, EU executives dedicated most of their messages to informing their audience about their activities (see results in the Online appendix). This is followed by tweets explaining why they exist and what they stand for. Only in third place do we observe messages pertaining to political outputs. With the official declaration of the state of pandemic in March 2020, the share of tweets informing the audience about executives’ output spikes to the first place and remains there for the next three months. In June, activity messages again catch up with output messages. These results suggest that the state of the pandemic conditioned the share of output tweets. Meanwhile, the executives traded activity messages for output messages; there is no substantial difference in the shares of other object types such as Identity and Mandate, Input seeking and Opinion. This might also relate to how the pandemic limited physical activities, providing less opportunities for EU executives to tweet about physical meetings and travels. Overall, while we see a change in the content that we expect to cue output legitimacy beliefs, the difference between content that we expect to cue different legitimacy beliefs remains relatively stable.

Next, we compare the content of the tweets across different types of executive actors, as presented in Figure 4. It is striking how there are strong similarities in the communication content between these different types of actors. This hints at a degree of centralisation of their communication. Our findings give the impression that EU executives may try to remain within the boundaries of some guidelines when conducting strategic communication on Twitter, in line with previous findings on European Commission communication strategies on social media as relatively centralised (Aagaard, 2023). However, this conjecture is not possible to prove or disprove with the limitations of our study.

Figure 4.

Share of the object of publicity in European Union (EU) executives’ messages.

Meanwhile, we also find several key differences in the content of the tweets of different types of executive actors. First and foremost, individual authorities such as presidents and commissioners tweet opinions more often than institutional accounts. Similarly, messages informing the audience about day-to-day activities are also more frequently sent by individuals rather than institutional accounts. One notable exception to this observation is the frequency of activity messages from the Commission and the ECB. These institutions often inform their audience about their day-to-day activities, including conferences, meetings, and workshops. Finally, replying to other users is more common among individual authorities such as commissioners. This implies that individual authorities are more engaged with their Twitter audience than other types of executives and might as such have stronger potential to cue positive input legitimacy beliefs.

One of our more surprising findings is the opinionated style of the EU executives’ communication. We observe that around 20% of the total number of tweets from EU executives express their opinions, position, and evaluation on issues while there is a considerable variation across different executive groups (see Figure 4).

As a single tweet might cue different legitimacy beliefs simultaneously, illustrated by the shares of the categories in Figures 3 and 4 summing up to more than 100%, one question remains open: How do they combine different content of communication in their tweets? To investigate possible patterns of combined messages, we conducted a cluster analysis. We do this for two reasons. First of all, there is no a priori theory that would suggest a particular pattern of combining various appeals in a public communication, therefore hypothesis testing methods such as regression analysis at this stage would not yield informative results. Secondly, this would allow us to obtain a more wholistic picture of how EU executives combine various contents in a single message without any learning bias. We conduct a logistic regression analysis to identify associations between executive type and the use of a particular combination once content combinations are identified.

To do this, we measured the similarity of tweets using the object of publicity indicators in a hierarchical clustering with the Ward method. The Ward method groups observations that present minimum variance as groups to form branches of the tree thus presenting more coherent groups (Ward, 1963). We utilised 22 different indices offered by Charrad et al. (2014) to identify the relevant number of clusters. We specifically aimed to determine a number of clusters between 1 and 7. This choice is informed by the number of categories for the object of publicity. On one extreme, the number of clusters could be 1 if executives combine all message types together in every tweet. On the other extreme, while unlikely, the number of clusters could be 7 if executives dedicate each tweet to only one type of message. Our analysis indicates that there are two clusters. The test results for determining the number of clusters are presented in the Online appendix.

The executives’ public communication messages cluster into two distinct groups. Most tweets, about 60%, fall in a cluster we call ‘opinionated communication’ tweets. All the categories of object of publicity we coded for are present in this cluster. As illustrated in Figure 5, tweets in this cluster mainly report on the day-to-day activities of the executives in combination with their opinions, outputs, as well as requests for input. The key difference between this cluster and the other one is the presence of opinion expressing messages. The other cluster containing ‘formal communication’ is dominated by output messages. It distinctly lacks any mention of opinions, replies, and more personal messages as tweets in the ‘opinionated communication’ cluster do.

Figure 5.

Message content composition of clusters.

To identify which group of executive actors are more likely to employ one of the two clusters, we fit a logistic regression model where the executive actor type is predictor, and the cluster membership of a tweet is the predicted value. Figure 6 below shows the odds ratio for each executive type to write an opinionated message. Since the sample consists of over 70.000 tweets, the significance threshold is set to five sigma level to so that type 1 error probability minimised.

Figure 6.

Odds ratio for executive actors’ use of opinionated messages.

Results from the regression model indicate that almost all the executives, with a slight deviation for DG institutions, are highly likely to employ opinionated communication over formal communication. Based on the odds ratio for each group, it is usually the actors in the limelight, such as presidents, vice-presidents and commissioners, that are most likely to employ this communication. The other types of executive actors such as director generals, their deputies and the DGs they manage are less likely to favour opinionated communication over formal communication compared to other executive actors in the sample.

We can draw several conclusions based on our observations of the content of executives’ messages. First and foremost, the communication of the more bureaucratic EU executives – agencies and DGs – has a stronger potential to cue positive output legitimacy beliefs than the more political EU executives, particularly the individual commissioners (including vice-presidents and presidents) (cf. Schmidt, 2020). This is indicated by a comparatively high frequency of output messages in their communication. This implies that their public communication practices, and attending legitimation potential, are mainly guided by a logic to “tell them what we do for them”. One possible explanation of this observation is that such bureaucratic executives rely on the technical knowledge of the subject they govern, i.e., technocratic legitimacy (Zürn, 2018), and that this in turn influences their communication. By focusing on the communication of their outputs, they may intend to reaffirm that they use their authority to execute the specific tasks delegated to them for the betterment of EU citizens. We know that the Commission has done so in the case of fiscal and economic supervision (Van der Veer and Haverland, 2018). This is also in line with DG COMM's aim to ‘focus on how the EU is responding to the crisis in order to save lives, alleviate economic impacts, and create new opportunities linked to the green and digital transitions’ (DG COMM, 2020: 4).

As indicated by the high frequency of opinionated communication messages, executives take risks by communicating their position on issues, asking for input, and engaging in a dialogue with their audience, albeit to a limited extent. This does not reflect a strategy of depoliticisation. Rather, such bold, outspoken messages create the risk of provoking opponents, thus generating backlash. But they may also create stronger engagement among supporters and other sympathetic audiences.

Discussion and conclusion

EU executive institutions and individuals have a precarious legitimacy relationship with their constituencies in the face of growing politicisation, external crises, and communication obstacles. Against this backdrop, social media platforms offer unique opportunities for EU executives to communicate with the public and foster their legitimacy beliefs. Our study has set out to investigate the trajectories of the EU executives’ public communication Twitter to cue positive legitimacy beliefs. We investigated this through manual and automated content analysis of over 70.000 tweets by 113 different EU executive accounts between December 1, 2019, and July 30, 2020.

While much has been written about the legitimacy of the EU, politicisation and self-legitimation by international organisations, few studies to date have engaged in an extensive and rigorous mapping of strategic communication on social media by the many individual and institutional social media accounts of the European Commission, EU agencies and other EU executive actors. This article contributes to the existing literature by conducting such an encompassing, rigorous mapping of strategic communication by EU executive actors. First, this reveals – in line with our expectations – that EU actors do indeed engage extensively in communication via social media like Twitter. We found a total of 113 different EU executive Twitter accounts, producing over 70.000 tweets in a period of eight months. Second, we found that these tweets contain appeals to various forms of legitimacy, including identity, input, throughput and output legitimacy. This variety matches our expectations, based on academic literature and official EU communication strategy documents.

Against our expectations, however, we found that EU executive strategic communication is rather opinionated and thus more ‘risky’ or ‘aggressive’ in its politicised environment. Following our initial finding that a variety of legitimacy appeals occur and that individual tweets frequently contain more than one of these appeals, we conducted a cluster analysis to distil patterns in this co-occurrence of legitimation. This analysis revealed that the most frequent type of tweet produced by EU executives contains a combination of appeals to output legitimacy, other forms of legitimacy and an opinion, often with a personal touch. This ‘opinionated communication’ style implies a rethink of EU executives as key actors in the politicisation of the EU. Where previous research characterises EU executives as risk-averse communicators, trying to depoliticise (or avert politicising) issues through technocratic communication focused on output (Pansardi and Tortola, 2022; Rauh et al., 2019), we showcase a more engaging EU executive. It goes beyond the aims of this paper to explain this phenomenon, but we allow ourselves to reflect on possible explanations. It is conceivable that the communication professionals operating the EU's socials enjoy more leeway than their colleagues responsible for official press releases. Where more official communication may have to go through more checks – including by legal services – perhaps social media officials enjoy more autonomy. It could also be that engagement in the form of likes and retweets provides EU executives with the incentives to go out of their comfort zone, or at least to seek its limits, a more common phenomenon in the realm of politics on social media (Bil-Jaruzelska and Monzer, 2022; Heidenreich et al., 2022). Future research could investigate this further, for example through interviews or participant observation with EU communication professionals.

Furthermore, we find that the EU executives largely refer to the EU, or the EU Commission, as a unitary entity, diffusing political responsibility. Such presentations of the EU as one actor, rather than emphasising the responsibilities of individual actors and institutions, might be harmful for its legitimacy. Their communication can result in de-legitimation rather than legitimation as the individual executives obfuscate their role in decisions and political outputs and point towards the polity in its entirety as the responsible actor.

To be able to reach these conclusions, we developed a novel analytical framework on potential trajectories for cueing legitimacy beliefs with strategic public communication. This framework synergises theoretical constructs from the literature on the legitimacy of international organisations with the literature on public relations. The framework allows us to conceptualise how different content of communication can be employed to appeal to four sources of legitimacy beliefs – identity, output, throughput, and input – and to explore this empirically. The synergy between EU legitimacy literature and public relations literature provides future research with tools to engage in further mapping and explanatory studies on how and why EU actors communicate the way they do.

Whereas this study fills an important gap with regard to our knowledge of how EU executives communicate on social media, it also leaves several questions unanswered. Our findings are surprising in terms of the legitimation strategies EU executives employ, which diverge from earlier studies of EU executives’ communication which have explored other arenas of communication. This raises questions about how the affordances of Twitter as a social media platform and the characteristics of the EU institutions shape such appeals to legitimacy beliefs. An avenue for further research to shed light on this, is to investigate the self-legitimation strategies of the EU comparatively, both across actors and communication platforms.

Further studies should also consider the temporal context. The COVID-19 pandemic was an extraordinary situation, which might have had consequences for EU communication, not only in terms of an intensification of the communication but also in its forms. While we to some extent examined this in our study, our finding that the initial phase of the pandemic did have a significant effect on the messages communicated underlines the importance of future studies exploring different periods in time, to see whether different situations have an impact on how the EU tries to legitimise itself on social media. As both the social media landscape – notably the take-over of Twitter by Elon Musk and change into X – and the geopolitical landscape in which the EU finds itself change, so could EU executive communication and attempts at self-legitimation more broadly.

The starting point of this article is EU executives’ need to build legitimacy in the eyes of the public. Our study documents patterns in EU executives’ public communication on Twitter. Yet, public communication only has the potential, and not a deterministic outcome, to influence the legitimacy beliefs of the audience. Future research would therefore do well to investigate how different communication content affects citizens’ legitimacy beliefs about the EU. Examining the effect of EU executives’ Twitter communication would greatly contribute to our understanding of the dynamics of sociological legitimacy.

Supplemental Material

sj-zip-1-eup-10.1177_14651165251320881 - Supplemental material for EU executives on social media: Assessing the potential and trajectories of legitimation via public communication on Twitter

Supplemental material, sj-zip-1-eup-10.1177_14651165251320881 for EU executives on social media: Assessing the potential and trajectories of legitimation via public communication on Twitter by Sina F Özdemir, Kristine Graneng and Pieter de Wilde in European Union Politics

Supplemental Material

sj-docx-2-eup-10.1177_14651165251320881 - Supplemental material for EU executives on social media: Assessing the potential and trajectories of legitimation via public communication on Twitter

Supplemental material, sj-docx-2-eup-10.1177_14651165251320881 for EU executives on social media: Assessing the potential and trajectories of legitimation via public communication on Twitter by Sina F Özdemir, Kristine Graneng and Pieter de Wilde in European Union Politics