What drives the focus of
European Parliament Members of
the 8
th
Legislature in posing
written oversight questions to
European agencies?
A quantitative and qualitative content analysis of
parliamentary written questions addressing
European agencies
Quirine Hoogeveen
S1376659
MSc. Public Administration: Economics & Governance
Leiden university
11-06-2019
Word count: 16105
Supervisor: Dr. E.M. Busuioc
1
Abstract
This thesis analyses the driving forces behind the parliamentary attention of the 8th European Parliament towards European agencies, in terms of written oversight questions. Through a quantitative and qualitative analysis of the written oversight questions posed by the Members of the European Parliament (MEPs) of the 8th Legislature, this thesis will make a contribution to the literature on incidental parliamentary oversight and the understanding of the EP’s attention towards the European agencies. In addition, this thesis highlights the deemed crucial matter of democratic control conducted by the European Parliament. This thesis applies a mixed research method: a quantitative and qualitative content analysis of the written questions posed by the MEPs. The quantitative content analysis has the principal objective to study whether a consistent pattern is to be found in the parliamentary focus in light of Font and Pérez-Durán (2016) findings that political salience and the number of personnel working at the agency determine the parliamentary focus. In addition, the quantitative content analysis sheds, in a suggestive way, light on the possible logics of the (inconsistent) patterns noticeable. The qualitative content analysis purposes to study in an explorative way the reputational focus pursued by the MEPs in posing their written oversight questions towards those European agencies. With regard to the drivers behind the MEP’s attention towards the European agencies, this thesis demonstrated that there is a partial consistent pattern to be found in terms of written questions and that MEPs direct their parliamentary oversight towards the performance reputational aspect of those European agencies positioned in the spotlight of the attention.
2
List of acronyms
AFSJ Area of Freedom Security and Justice
ESAs European Supervisory Authorities
EP European Parliament
EU European Union
MEPs Members of European Parliament
3
P. List of Figures:
Figure 1. The accountability mechanisms available to the European Parliament 18
Figure 2. The rise of the European agencies 32
Figure 3. Number of individuals employed by the European agencies 38
Figure 4. The substantive by MEPs towards Europol in posing written 44 oversight questions following the four dimensions of
organisational reputation theory
Figure 5. The substantive by MEPs towards Frontex in posing written 45 oversight questions following the four dimensions of
organisational reputation theory
List of Tables:
Table 1. The four dimensions of organisational reputation theory 16
Table 2. The four dimensions of organisational reputation theory 27 applied to the written oversight questions
Table 3. The results on parliamentary attention towards the European agencies 34 in posing written questions when searching on the acronym in text and/or title
Table 4. Number of written questions per year that named Europol 42
Table 5. Number of written questions per year that named Frontex 43
Table 6a. The reputational focus by MEPs towards Europol in posing written 44 oversight questions
Table 6b. The substantive by MEPs towards Europol in posing written 44 oversight questions following the four dimensions of
organisational reputation theory
Table 7a. The reputational focus by MEPs towards Frontex in posing written 45 oversight questions
Table 7b. The substantive by MEPs towards Frontex in posing written 45 oversight questions following the four dimensions of
organisational reputation theory
Table 8. The reputational focus of European agencies, following their specialisation 47
Annex:
Annex I European Agencies 58
Annex II Full name in text and/or title 59
Annex III The 24 printed international newspapers used by 60 Font and Pérez-Durán (2016) in their analysis on
the political salience of European agencies
Annex IV The political salience of European agencies 61
Annex V Staff number of European agencies 62
4
Table of Content
1. Introduction ... 6
1.1 European agencies: the problem of its raison d’être ... 6
1.2 The democratic control of European agencies by the European Parliament ... 7
1.3 Thesis contribution: a quantitative and qualitative content analysis of the parliamentary oversight towards European agencies via written questions ... 8
1.4 Research question ... 10
1.5 Thesis outline ... 10
2. Theoretical framework ... 11
2.1 Conceptualization ... 11
2.1.1 Defining ‘accountability’ ... 11
2.1.2 A terminology for ‘European agencies’ ... 12
2.1.3 Agencies’ independence, autonomy and accountability ... 12
2.2 Theoretical approaches towards accountability and multiple audiences’ expectations.. 13
2.2.1 ‘The forum versus the principal’ ... 13
2.2.2 Linking fire-alarm oversight to the ‘drifting’ forums ... 15
2.2.3 Managing organisational reputation in web of audiences ... 15
2.3 Literature review ... 17
2.3.1 The rise of the EP’s accountability power ... 17
2.3.2 Parliamentary oversight towards European agencies via written questions ... 18
2.4 Thesis’ expectations ... 20
2.4.1 Expectation I (quantitative content analysis) ... 20
2.4.2 Expectation II (qualitative content analysis) ... 21
3. Research Methodology ... 23
3.1 Research method: a quantitative and qualitative content analysis ... 23
3.2 Operationalization of the 2-stage design. ... 23
3.2.1 Quantitative content analysis: the parliamentary attention towards European agencies in posing written questions ... 23
3.2.2 Qualitative content analysis: the reputational focus MEPs pursue in posing written oversight questions towards European agencies ... 25
3.3 Validity and reliability of the mixed content analysis ... 29
3.4 Method of data collection ... 30
4. Emperical analysis ... 31
4.1 The ‘agencification’ phenomenon – a family on its rise ... 31
4.2 Quantitative content analysis ... 32
4.2.1 The parliamentary focus of the 8th Legislature ... 32
5
4.2.3 Consistent patterns in parliamentary attention towards agency’s staff number? ... 36
4.2.4 Quantitative content analysis – results ... 39
4.3 Qualitative content analysis ... 40
4.3.1 Introducing the cases under analysis: Europol and Frontex ... 40
4.3.2 The parliamentary attention towards Europol and Frontex in terms of written oversight questions ... 42
4.3.3 Qualitative content analysis - results ... 43
5. Conclusion ... 48
5.1 Concluding remarks on the parliamentary focus by MEPS of the 8th Legislature towards Europol and Frontex ... 48
5.2 Research limitations ... 50
5.3 Suggestions for further research ... 50
7. Works cited ... 51
6
1. Introduction
1.1 European agencies: the problem of its raison d’être
The European Union (EU) has witnessed the rise of European agencies, which are deemed as important actors inside the institutional regulatory domain of the EU (Groenleer, 2009; Busuioc & Rimkuté, 2019). The European agencies vary in their institutional format from other EU organisations, as they are non-majoritarian in nature, operate (semi-)autonomously and execute tasks delegated to them (Busuioc, 2013a). One of the principal justifications for the establishment of the agencies, is the institutionally independent nature of the bodies. As the agencies are institutionally on their own, they are not directly controlled by politicians who might have their aims set on electoral returns (Majone, 1997). When an organisation is not directly part of the hierarchical chain, it could secure credibility in the policies pursued. In addition, independence could foster expertise. As explained by Majone (1997) “faith in the power of expertise as an engine of social improvement – technical expertise which neither legislators, courts nor bureaucratic generalists presumably possess – has always been an important source of legitimisation for regulators” (p. 152). The European Commission highlighted the importance and benefits of the institutionally independent characteristics as well of European agencies by stating that “the independence of their technical and/or scientific assessments is, in fact, their real raison d’être. The main advantage of using the agencies is that their decisions are based on purely technical evaluations of very high quality and are not influenced by political or contingent considerations” (European Commission, 2002, p. 5). Essential to underline is that, the advantages and the perceived legitimacy of the agencies’ institutional character, come forth out of the regulatory domain in which the agencies operate (Majone, 1997). Policy areas that for instance involve economic redistributions are perceived as belonging to directly elected public officials. The execution of these tasks is not perceived as legitimate actions of non-majoritarian bodies.
When observing the growth in the number of European agencies, it almost appears as if the EU aimed to address every new emergency (varying from food safety issues to unsafe planes) with the creation of (yet again) a new agency (Van Ooik, 2005). However, the very principal reason why the European agencies were being ‘advertised’ as the right actors in their field, i.e. their independence, became a widely disputed characteristic of the agency themselves. “The once heralded ‘solution’ became a ‘problem’ in its own right” (Busuioc, 2013a, p. 4). The fact that agencies act independently and operate in a rather complex institutional framework, combined with the important responsibilities delegated to them, positioned the agencies’
7
accountability and democratic legitimacy in the centre of discussion (see Everson, 1995; Curtin, 2005; Wonka & Rittberger, 2011, Busuioc, 2013a). In consequence, the perceived problem focuses on the question of how to reconcile the agencies’ independence with the possibility to hold the agency to account and to supervise their actions (Busuioc, 2013a, Groenleer, 2006). As the agencies have vested a prominent place inside the EU institutional framework, both in terms of quantity as the importance and sensitivity of the responsibilities carried out, the question of the agencies’ accountability is perceived as a crucial matter of concern.
1.2 The democratic control of European agencies by the European Parliament
Despite the fact that there are distinctive accountability mechanism in place to scrutinize the agencies (see Groenleer, 2009; Busuioc, 2013a), the Commission, the Council of the EU and the European Parliament (EP) identified in their Common Approach (2012) for the European agencies that the agencies’ accountability is one of the objectives further to be met (European Parliament, Council of the EU & European Commission 2012a; European Parliament, Council of the EU & European Commission 2012b). The accountability power of the EP over the European agencies was initially rather limited (Busuioc, 2013a). However, following the Treaty of Lisbon, the EP gained enhanced powers towards European agencies (see Busuioc, 2013a; Lord, 2011; Groenleer, 2009; Bach & Fleischer, 2012; Jacobs, 2014). In consequence, the EP has currently a variety of distinctive mechanism at its disposal to scrutinize the agencies, which focus both on the performance as the budget of the European agencies (Busuioc, 2013a; Groenleer, 2009). When reflecting on the aforementioned discussion on the European agencies’ accountability and democratic legitimacy, it could be argued that the oversight carried out by the EP is of special relevance regarding the democratic control over the European agencies (Busuioc, 2013a; Lord, 2011). This rhetoric has as well been put forward by the EP itself;
“The establishment of parliamentary control over the structure and the work of the regulatory agencies is consistent with the classic democratic principle requiring political responsibility of any body yielding executive power. The possibility of the European Parliament assigning political responsibility to the agencies concerned touches on a core principle of representative democracy, which consists in examining the legality and expediency of the choices made by the executive power” (European Parliament, 2008).
8
1.3 Thesis contribution: a quantitative and qualitative content analysis of the parliamentary oversight towards European agencies via written questions
While a hard-won institutional privilege, surprisingly the academic literature on parliamentary oversight towards European and national agencies demonstrates that the parliamentary attention varies (see Dudley, 1994; Hogwood, Judge and McVicar, 2000; Pollitt, 2006; Koop 2011; Busuioc 2013a). The authors empirically concluded that parliamentary oversight is incidental, often triggered by the political salience of the agencies. As put in the words of Busuioc (2013a) “attention is reportedly targeted on highly mediatized and politicized issues. By remaining confined only to certain aspects, accountability tends to be a bit patchy and not geared towards the ‘broad picture’” (p. 133). Following this line of reasoning, the accountability behaviour by the EP is termed as ‘fire-alarm’ oversight instead of ‘police-patrol’ (Busuioc, 2013a; McCubbins & Schwartz, 1984). Whereas police-patrol oversight is carried out in an ongoing, centralized and direct way, fire-alarm oversight is conducted in an incidental and decentralized manner, involving distinctive actors (McCubbins & Schwartz, 1984; Zegart & Quinn, 2011). Legislators might conduct more fire alarm oversight, as this ensures that public officials focus on those issues where their crucial constituents care about (Zegart & Quinn, 2011). This is deemed essential for the politicians in light of their electoral returns. An occasional EP oversight means at the same time, that neglect in parliamentary attention might be visible, especially in areas that are not subject to high pubic attention (Busuioc, 2013a; Schillemans & Busuioc, 2014).
Through an analysis of written oversight question posed by the Members of the European Parliament (MEPs), this thesis will make a contribution to the observed variation in parliamentary oversight by academics and the understanding of the EP’s attention towards the European agencies. Written questions function as a medium for the MEPs to acquire information and to put perceived problems under the eye of the Council and Commission (Raunio, 1996). In terms of empirical research purposes, recorded parliamentary questions are indicative of the preferences, priorities and role perceptions of the MEPs (Martin, 2011). Therefore, an analysis of the publicly available written questions provides the opportunity to study the incidental parliamentary attention and the substantive focus MEPs pursue in conducting their parliamentary oversight. Via a content analysis of the written questions addressing European agencies, this thesis will study both quantitatively as qualitatively the variation in parliamentary attention towards the agencies. While there has been quantitative research on the parliamentary oversight towards European agencies (see Font & Pérez-Durán, 2016), the literature on parliamentary oversight lacks a qualitative analysis of the substantive
9
focus MEPs pursue in posing their written questions towards European agencies. This combined research method generates the possibility to analyse what aspects of agencies’ activities truly matter for the EP’s parliamentary oversight.
The quantitative content analysis has the principal objective to study which European agencies received most parliamentary attention in terms of written questions during the 8th parliamentary term (2014 – 2019). It will built further on the empirical conclusions set forward in academic literature that political salience matters for parliamentary oversight and will pay specific attention to the findings of Font and Pérez-Durán (2016). These authors analysed quantitatively the written oversight questions posed by MEPs during the parliamentary term of 2009 – 2014 and demonstrated that MEPs asked more written oversight questions to European agencies that are politically salient or have relatively large capacity in terms of staff. In this thesis, the results of the quantitative content analysis will provide the basis for (feed in) the qualitative content analysis, where the attention is directed at the substantive focus of MEPs in posing written oversight questions towards European agencies. Building forward on the notion that an European agency has accountability relationships with distinctive forums, as written down in the analytical definition of accountability by Bovens (2007), European agencies are presumed to find themselves in a web of different audiences, all having their own expectations regarding the agencies’ behaviour (Busuioc & Rimkuté, 2019). According to organisational reputation theory, the manner an agency copes with these expectations, has implications for their reputation and in consequence their discretion (Carpenter, 2010; Carpenter & Krause, 2011; Busuioc & Rimkuté, 2019). Following this rhetoric, the substantive focus MEPs pursue in their written questions is studied in this thesis, according to the four reputational dimensions of Carpenter (2010). This creates the opportunity to study the focus of an accountability forum on which these reputational dimensions tend to pertain to. In consequence, by quantitatively and qualitatively analysing the EP as an accountability forum in relation towards the European agencies, this thesis makes a contribution to the empirical and theoretical understanding of variation in parliamentary oversight and highlights the deemed crucial matter of democratic control over the European agencies by the European legislators.
10
1.4 Research question
1.5 Thesis outline
In answering the research question, this thesis is guided by the following structure. Chapter 2 exists out of the Theoretical Framework, which starts with a clarification of the principal concepts under analysis: ‘accountability’, ‘European agencies’ and ‘autonomy’. Subsequently, the widely applied and criticized theoretical framework with regard to accountability, the principal-agent theory, is being explained. The identified gaps in the principal-agent approach are ‘covered’ by the fire-alarm oversight theory and reasonings stemming from organisational reputation theory. These theoretical insights are followed by an overview of the perspectives taken by academics towards the EP’s accountability power and the parliamentary oversight to European agencies. The Theoretical Framework ends with two expectations. The Research Methodology in chapter 3 explains that these two formulated expectations will be analysed via a combined research method: a quantitative and qualitative content analysis. It clarifies the operational steps being taken, the validity and reliability of the mixed content analysis, and the method of data collection. In chapter 4 the Empirical Analysis is carried out. The empirical research starts with the quantitative content analysis of the total written questions addressing the European agencies. The quantitative content analysis determines subsequently the case selection for the qualitative content analysis, where the substantive focus of the written oversight questions is being studied. In chapter 5 concluding remarks on the parliamentary focus are formulated. The Conclusion includes acknowledgements of the research limitations as well and describes suggestions for possible further research.
What drives the MEPs focus in posing written oversight questions to European agencies? o Which agencies trigger the parliamentary attention of MEPs when posing written
questions? (quantitative content analysis)
o Which reputational focus do MEPs pursue in posing written oversight questions towards European agencies? (qualitative content analysis)
11
2. Theoretical framework
2.1 Conceptualization
2.1.1 Defining ‘accountability’
The term ‘accountability’ is commonly used by policymakers (Bovens, 2007) and the literature on public administration has found an agreement in defining it a “complex and chameleon-like term” (Mulgan, 2000, p. 555). While acknowledging that accountability is an equivocal concept and distinctive meanings are attached to the term, this thesis will follow the definition of accountability as put forward by Bovens (2007). According to Bovens (2007), accountability entails “a relationship between an actor and a forum, in which the actor has an
obligation to explain and justify his or her conduct, the forum can pose questions and pass judgements, and the actor may face consequences” (p. 450). The ‘forum’ can be a principal
who directly mandates a responsibility, or it can be a public organisation/official maintaining a more indirect accountability relationship with the actor. The ‘actor’ could be a collective entity or an individual person. The ‘obligation’ to render account on the side of the actor, might follow official procedures, but can as well be of informal nature. In the account giving, the actor needs to provide information to the forum on its behaviour. The forum should be able to scrutinize and ask questions about the activities of the actor. The essential objective hereby, is to receive an all-encompassing view on the appropriateness of the information provision by the actor and the rightness of its actions, so the forum can make a decision. The ‘consequences’ that might be charged up on the actor can be punitive or can signal appraisal.
This thesis will follow Bovens’ (2007) definition of accountability, as the author puts the concept of accountability in an analytical scope, as oppose to depicting it as an evaluative standard. In addition, this definition of accountability recognizes that distinctive accountholders (in addition to the principal) are present, which is of relevance in light of this thesis’ research objective. Accountability is in this way perceived as an ex post process, as oversight is conducted after the responsibility has been delegated (Busuioc & Groenleer, 2013). In order to not fall victim to (common) misunderstandings of the word ‘accountability’, Bovens (2007) aimed to clarify three misinterpretations, which will be repeated here. Firstly, accountability should not be confused with the term ‘transparency’; whereas the scrutiny of an actor is present in the first concept, it is absent in the latter. Secondly, accountability should not be misinterpreted as ‘inclusion’ or ‘being receptive to distinctive visions,’ as the latter notions miss
12
the rationale for (1) why particular actions are chosen (2) how they are being judged and (3) which consecutive actions might be undertaken. Finally, accountability should not be wrongly mistaken for ‘control’; accountability is a manner to pursue control, however exercising control does not come with accountability behaviour at all times.
2.1.2 A terminology for ‘European agencies’
Public officials and academia do not consistently use one definition to describe an ‘European agency’ (Groenleer, 2009; Busuioc, 2013a). Despite this lack of a shared terminology used, the European agencies do possess unique characteristics. Based on these shared features, Busuioc (2013a) defined EU agencies as “specialized, non-majoritarian
bodies, established by secondary legislation, which exercise public authority and are institutionally separate from the EU institutions and are endowed with legal personality” (p.
21). A closer look is taken at what the components of this definition insinuate. Agencies are set up with a specific purpose and can therefore be seen as specialists in their field (Busuioc, 2013a). With respect to their institutional structure, ‘non-majoritarian’ refers to the notion that the public officials working in the agencies are not being elected by the public citizens (Majone, 1998; Groenleer, 2009). Being created on the basis of secondary legislation implies that the agencies “are not statutorily independent” (Pollitt et al, 2004, p. 8). The fact that agencies are endorsed to follow statues which have been democratically agreed upon, is one of the reasons why agencies are perceived as possessing procedural legitimacy (Majone, 1997). The definition of Busuioc (2013a) outlined as well that European agencies carry out public authority. The European agencies are public bodies following public law and are perceived as acting in the EU’s citizens interest (Busuioc, 2013a, Pollitt et al, 2004; “Decentralised Agencies”, 2019). The institutional separation means that the agencies are not directly part of a hierarchical chain (Groenleer, 2009); they possess an “identifiable, separate, organisational structure with [an] own name” (Talbot, 2004, p. 8). Finally, the European agencies have a legal personality, which implies that the agencies can in a formal way act independently (Busuioc, 2013a; Groenleer, 2009). Due to their legal personality, European agencies could for instance settle agreements with other actors (Groenleer, 2009).
2.1.3 Agencies’ independence, autonomy and accountability
The introduction of this thesis stressed that the institutional independence is the principal raison d’être for European agencies. However, following the rhetoric of Groenleer
13
(2009), the institutional set up of European agencies might better be defined by the word ‘autonomy’ than ‘independence.’ Although both concepts appear at first sight similar, they could mean different things in practice. Whereas ‘independence’ insinuates a “condition of being politically free”, ‘autonomy’ refers to a “capacity to manage one’s own affairs” (Groenleer, 2009, p. 29). Following from this, a European agency possesses autonomy, yet is not independent (Groenleer, 2009). European agencies can to an extend decide upon their own actions, however in the end they remain under the supervision of other public officials. Autonomy can be separated in de jure (formally laid down) autonomy and de facto (the true) autonomy (Busuioc & Groenleer, 2013). Simply put, formal accountability determines the official autonomy ‘box’ (be it concrete or vague), and the actual autonomy follows from the manner public organisations ‘fill in’ this box (Groenleer, 2009). As put forward by Groenleer (2009), “the actual autonomy refers to an EU agency’s ability to choose among different behaviours pertaining to what its critical tasks are, and how and with what resources it performs these tasks, and how the agency, in the performance of its tasks, relates to other actors in its environment” (p. 39). Based on this, the actual autonomy is determined by both the organisations’ internal interpretations and the way it is being approached by external actors (Busuioc & Groenleer, 2013). What kind of actual autonomy the European agencies develop, could vary per agency (Groenleer, 2009). Autonomy and accountability, as defined above, are related because an ex post oversight mechanism becomes of crucial importance when a public body is autonomous and not directly part of the hierarchy (Busuioc & Groenleer, 2013). “It is autonomy that renders accountability pertinent” (Busuioc & Groenleer, 2013, p. 288).
2.2 Theoretical approaches towards accountability and multiple audiences’ expectations
2.2.1 ‘The forum versus the principal’
With respect to its assumptions regarding the process of accountability, the theoretical framework of (rational) principal-agent has been widely applied within the academic disciplines of public administration, political science and International Relations (Schillemans, 2013). The principal-agent theory specifies that a principal empowers an agent to carry out a responsibility (Schillemans & Busuioc, 2014). The principal attaches importance to the delegated activity and prefers this action to be efficiently fulfilled. However, the interest of the principal and the agent might not be in line with each other, as they might pursue distinctive goals. This divergence
14
might have the effect that implementation of the delegated task ‘drifts away’ from the original intentions of the principal. To secure that the agent follows the pathway as desired by the principal, distinctive ex post and ex ante control mechanisms are initiated and subsequently instituted in a contractual format. In this respect, the principal aims to foster account-giving by the agent following the established accountability institutions. In addition, the principal prefers to correct the agent’s behaviour where necessary. However, the agent might circumvent the radar of control by using information asymmetries to its advantage. Therefore, the central challenges in the accountability processes according to the principal-agent framework are the attempts to secure that the agent stays on the right path and the endeavours to foster corrections.
The principal-agent theory has not only been extensively used within the three academic fields, it has as well been widely disputed for their assumptions regarding accountability (Schillemans, 2013). Schillemans and Busuioc (2014) demonstrated that the principal-agent rhetoric does not provide an accurate depiction of the accountability relationships and provided “a warning against simply taking over general theoretical assumptions from the P-A model and applying them indiscriminately to accountability research” (Schillemans & Busuioc, 2014, p. 210-211). In their study, Schillemans and Busuioc (2014) applied the definition of Bovens (2007) and stressed the rhetoric of distinctive accountholders. Their central argument rested on the observation that the ‘drifting’ in the accountability relationship does not prevail on the side of the agent; it is the forum that frequently does not take on their responsibilities and obligations, and therefore fails. As described by Schillemans and Busuioc (2014), the forum falls short in the sense of “not caring about the delegated tasks, not wanting to hold their agents accountable,
drifting away in preferences from what has been formally instituted, failing to take
accountability information seriously and failing to correct and redress the behaviour of agents when feasible” (p. 205). The forum therefore does not conduct accountability behaviour as actively and as consistently as the principal in principal-agent theory would do. One explanation Schillemans and Busuioc (2014) put forward with respect to the ‘wandering’ forum, stems from the fact that a forum is not part of the accountability relationship; it has not mandated the actor with power or resources which originally belong to the forum. This marks the stark difference between a ‘principal’ and a ‘forum,’ as became apparent in the definition of Bovens (2007). Whereas the first is ‘personally’ involved in the delegation process, the latter is an ‘outsider’ watching (or not even). As the delegation might not be of immediate advantage to the forum, it might not be particularly keen to conduct oversight functions. In consequence, the forum might only be interested in exercising accountability practices when they are encouraged to do so, that
15
is “when there are particular incentives present in that direction” (Schillemans & Busuioc, 2014, p. 207)
2.2.2 Linking fire-alarm oversight to the ‘drifting’ forums
McCubbins and Schwartz (1984) outlined two models of parliamentary oversight: police-patrol and fire-alarm. Police-patrol has two defining features: detection occurs continuously and the oversight is carried out in a centralized manner (McCubbins & Schwartz, 1984; Zegart & Quinn, 2011). Instead of checking the radar repeatedly, fire-alarm oversight presupposes that scrutiny only occurs “when aggrieved parties really care” (Zegart & Quinn, 2011, p. 62). In addition, fire-alarm oversight involves indirect supervision conducted by multiple parties (McCubbins & Schwartz, 1984; Zegart & Quinn, 2011). In their study of congressional oversight, McCubbins and Schwartz (1984) argued that the congress pursued more fire-alarm oversight than police-patrol and hereby focused on those political matters which their ‘crucial’ voters care about (McCubbins & Schwartz, 1984; Zegart & Quinn, 2011). The worries of these people are deemed as most essential to the public officials with an eye on their re-election. As explained by Zegart & Quinn (2011), “citizens who are concerned and organized enough to ring an alarm are citizens who are likely to get out the vote and provide other campaign support” (p. 62). The theoretical insight of fire-alarm oversight can be linked to the remarked deficit in the principal-agent theory of drifting forums. Both share the assumption that the parliamentary oversight is conducted in an incidental and indirect way, as oppose to a systematic scrutiny.
2.2.3 Managing organisational reputation in web of audiences
The (possible) explanation for forum drifting, as defined above, and the theoretical insight of fire-alarm stress the presence of multiple accountholders with respect to an agent. The agent can therefore be portrayed as an entity that is being watched (incidentally) by distinctive audiences, who all have their own expectations regarding the organisations’ behaviour (Busuioc & Rimkuté, 2019). The manner the agent acts on these expectations has implications for their reputation. Carpenter (2010) stressed in his chapter on “Reputation and Regulatory Power” (p.33), the importance of audiences for regulators specifically, by applying a reputation-based outlook. According to Carpenter (2010), “reputations are composed of symbolic beliefs about an organisation – its capacities, intentions, history, mission – and these images are embedded in a network of multiple audiences (p. 33). Concretely, the influence of
16
audience on regulation is twofold. Firstly, audiences can strengthen or weaken the power of a regulator, by considering the regulator an authority (or not). In this vein, developing an organisational reputation that is desired, might create public appraisal, could fuel the regulator’s autonomy and/or gives them leverage during waves of opposition (Busuioc & Rimkuté, 2019). Secondly, the regulator aims to adjust their conduct to the specific audience’ perceptions (Carpenter, 2010). Therefore, developing an organisational reputation can be perceived as a ‘management’ (Busuioc & Rimkuté, 2019) in which the organisation aims to cultivate a ‘niche’ (Busuioc & Lodge, 2015) and strategically protects itself (Rimkuté, 2018). As became apparent in the definition as put forward by Carpenter (2010), organizational reputation is not a concept which is formally written down; it is an “organizational image” (p. 46). According to Carpenter (2010), this reputational image has four dimensions: performative, moral, technical and legal, which are defined in the Table 1 below.
Carpenter and Krause (2011) explained that it is extremely difficult for a public organisation to realize each of the four organizational reputational dimensions, as the distinctive dimensions might stand in contradiction with eachother. An organisation could for instance execute their task in a highly efficient manner (i.e. meets the performative dimension of reputation), yet is not transparent in the means they apply to reach their end (i.e. lacks the moral dimension of reputation) or does not adhere to the formally laid down rules (i.e. an absent legal-procedural dimension of reputation) (Carpenter, 2010). The organisation therefore needs to focus on a perceived ‘priority’ dimension, rather than aiming to fulfil all four dimensions. Based on the four reputational dimensions of Carpenter (2010), Busuioc & Rimkuté (2019) formulated distinctive expectations regarding the agencies’ reputation building. The authors expected that agencies would predominantly pursue a technical reputation, as their expertise is the principal reason why the agencies have been established and why they are perceived as legitimated actors
Performative dimension of reputation Focuses on the way an organization carries out its assigned responsibilities
Moral dimension of reputation Centres on the moral and ethical aspect of the organisation’s behaviour
Technical dimension of reputation Highlights the technical (independent) expertise of the organisation
Legal-procedural dimension of reputation
Emphasizes the legal basis of the organisation’s actions
Table 1. The four dimensions of organisational reputation theory (Carpenter, 2010; Carpenter & Krause, 2011)
17
in the regulatory field. In addition, following the rhetoric put forward by Majone (1998) that technique is detrimental for the performance delivered, Busuioc & Rimkuté (2019) stressed that the technical dimensions might be accompanied by the performative dimension over time. The authors predicted that the agencies would give least priority to the moral dimension, as this reputational dimension has a political undertone which stands in stark contrast with the desired agencies’ independence in carrying out their expertise.
2.3 Literature review
2.3.1 The rise of the EP’s accountability power
The Treaty of Rome (1957) specified that the EP has a consultative function in the decision-making process, as currently laid down in Article 289 Treaty on the Functioning of the EU (TFEU) (Servent, 2018, “Consolidated Version of the TFEU”, 2012). This decision-making procedure implied that the Commission would submit a proposal, which had to be adopted by the Council. However, before making its decision, the Council needed to be consulted by the EP. As this consultation was of non-binding nature, the Commission and the Council were in practice the principal powers. However, the Treaty of Lisbon (2007) termed the newly gained co-decision role of the EP the “ordinary legislative procedure”, as written down in Article 294 TFEU “Consolidated Version of the TFEU”, 2012). This clause implied that the EP would be in principal on equal foot with the Council regarding the co-decision function (Servent, 2018). Following its new role, the EP could determine that oversight clauses would be incorporated in the agencies’ regulations (Busuioc, 2013a). The principal effect of this was that the EP had more opportunity to scrutinize the European agencies.
The EP has currently distinctive mechanism at its disposal to hold the European agencies to account (Busuioc, 2013a; Groenleer, 2009). Broadly speaking, the EP’s accountability instrument can be divided in two categories (see examples of instruments in Figure 1 below). The first group of accountability instruments focuses on the performance of the agencies and how the agencies carry out the responsibilities delegated to them. This form of oversight is for instance visible in the parliamentary hearings or the EP’s evaluation of the agencies’ annual reports. Following the EP’s budgetary and discharge authority, the second category of mechanisms centres on the financing of the agencies’ budget. Based on these powers, the EP can carry out substantive parliamentary oversight towards partial or completely EU funded
18
agencies (Busuioc, 2013a; Jacobs, 2014; Groenleer, 2009; Bach & Fleischer, 2012; Lord, 2011). One of the financial instruments available to the EP, which the agencies perceive as an impediment to their autonomy, is the EP’s influence over the provision of financial reserves (Groenleer, 2009). According to article 23 and 24 of the Financial Regulation, the EP’s Committee on the Budgets can decide to not make reserves available to the agency, until it is of the opinion that the required criteria are met (European Commission, 2018; Jacobs, 2014; Groenleer, 2009). One of these established prerequisites is for instance that the specialized committee to which the agency belongs, needs to give their endorsement to the agency’s work program and capacity, before reserves are made available (Groenleer, 2009; Busuioc, 2013a).
2.3.2 Parliamentary oversight towards European agencies via written questions
The academic literature on parliamentary oversight towards European and national agencies showed that the parliamentary attention occurs in a rather incidental way (see Dudley, 1994; Hogwood, Judge and McVicar, 2000; Pollitt, 2006; Koop 2011; Busuioc 2013a). The authors demonstrated that the political salience of agencies attracts the attention of the legislators. The concept of political salience refers to the cruciality an issue receives (Warntjen, 2011). The perception of viewing a matter as pivotal has different ‘sources’, as it can stem from “its (estimated) policy impact, the political sensitivity of an issue or the attention it receive” (Warntjen, 2011, p. 169). The rational for a focus on political salience by legislators, might be that public officials prefer to dedicate their scare time and capacities more to agencies dealing
EP's accountability power focused on the peformance of EU agencies (mainly by EP's specialized committees)
•Representation in the management board •Parliamentary hearings
•Receives annual work programs
•Gives advice on multiannual work program •Written questions
•Visits to the agency
EP's accountability power focused on the budget of EU agencies
(mainly together with EP's Committee on Budgets or EP's Committee on Budgetary Control) •Budgetary power
•Discharge power (over EU funded agencies)
Figure 1. The accountability mechanisms available to the European Parliament (see Busuioc, 2013a; Groenleer, 2009; Jacobs, 2014)
19
with issues of public concern, than to those who circumvent the public radar (Koop, 2011). This might either come from the fact that the politicians have a sincere interest in the matter, or due to electoral considerations. The central implication of the empirical observation that legislators devote their attention to political salient agencies, is that some agencies do receive considerably oversight, whereas others are neglected (Koop, 2011; Busuioc, 2013a; Schillemans & Busuioc, 2015). This might give rise to concerns on the democratic control and the adequacy of the EP’s accountability behaviour (Koop, 2011; Busuioc, 2013a).
As visible in Figure 1, an instrument the EP can use to scrutinize European agencies, is the formulation of written questions. Rule 130 of the “Rules of Procedure” of the European Parliament (2014-2019) stipulates that “a Member, a political group or a committee may put questions for written answer to the President of the European Council, to the Council, to the Commission or to the Vice-President of the Commission / High Representative of the Union for Foreign Affairs and Security Policy” (European Parliament, 2019). Individually or collectively, the MEPs can pose at maximum twenty questions during a time span of three months. In comparison with oral questions and questions posed during scheduled question time, the possibility to ask written questions is most frequently used by the MEPs (Raunio, 1996). The underlying reason for this might be that the for the formulation of a written question, less procedures need to be followed (Jensen, Proksch & Slapin, 2013). When MEPs adhere to the Rules of Procedure, they are allowed to pose a question whenever they feel like doing so and on whatever issues they prefer to be clarified (more). With respect to the questions posed by legislators, Martin (2011) considered that this instrument possesses “the potential to be an important ex post control mechanism” (p. 265). However, as become clear from the EP’s procedural rules, the MEPs are not allowed to directly forward a written question to the European agencies themselves. The formulation of written questions can therefore be perceived as an indirect form of parliamentary oversight.
In light of the written oversight questions posed towards European agencies, Font and Pérez-Durán (2016) purposed to explain what the driving forces were behind the questions formulated by MEPs during the 7th Legislature. In their quantitative study, the authors analysed the written questions posed by legislators having a seat in the EP during the period 2009-2014 and studied both relevant legislative (MEPs’) as executive (agencies’) characteristics. With respect to the agencies’ aspects of influence, Font and Pérez-Durán (2016) demonstrated that during the parliamentary term of 2009-2014, the legislative oversight was mostly triggered by European agencies that were politically salient (in terms of public attention received in printed media) and prompted by those agencies employing a relatively large number of personnel. With
20
regard to an explanation for the relation between parliamentary attention and the agencies’ political salience, Font and Pérez-Durán (2016) built forward on the rhetoric as defined above. They suggested that MEPs focus on those agencies widely discussed by the European public, as the legislators formulated a more prevailing political view/interest on this matter. As regards the agency’s number of personnel, Font and Pérez-Durán (2016) proposed that MEP’s might devote more oversight to the bigger agencies, as these have larger organisational capacities and carry out more substantial responsibilities.
2.4 Thesis’ expectations
2.4.1 Expectation I (quantitative content analysis)
By applying Boven’s (2007) analytical definition of accountability centred on multiple accountholders and by pursuing the understanding of ‘wandering’ forum and fire-alarm oversight, this thesis will follow the theoretical insight that the EP is an accountability forum, incidentally using its available accountability mechanisms. This can be linked to the identified deficit in the principal-agent theoretical framework, the most widely applied (and most disputed) approach to accountability processes in the discipline of public administration. Since this theoretical approach perceives a direct relationship between the accountholder and account-giver and presumes that accountability problems prevail due to a ‘drifting agent’, it cannot shed light on an indirect and neglecting parliamentary oversight. Following the empirical observations on the non-systematic nature of parliamentary oversight, this thesis anticipates that the EP conducts accountability practices towards the European agencies in an incidental way, that might be driven by electoral incentives or inducive political interests by the legislator. Building principally forward on the study of Font and Peréz-Durán (2016) on the 2009-2014 Legislature, this thesis formulates the expectation that MEPs of the 8th parliamentary term (2014-2019) will pursue a continuing focus on European agencies that are politically salient and relative large in terms of staff number when formulating their written questions. This section of thesis hereby maintains the principal objective to analyse whether the patterns in parliamentary focus towards the European agencies of the 2014-2019 MEPs, are consistent with the patterns in parliamentary attention of the 2009-2014 legislators, as empirically tested by Font and Pérez-Durán (2016). The European agencies receiving the attention will inform the
21
case selection for the qualitative content analysis of the substantive focused pursued by the MEPs in asking written oversight questions.
2.4.2 Expectation II (qualitative content analysis)
Taking organisational reputation theory as guidance (Carpenter, 2010; Carpenter and Krause, 2011), Busuioc and Rimkuté (2019) argued that European agencies might pursue a principal focus on the technical dimension. As the agencies become older, this focus can be accompanied by an attention on the performative dimension. Since the moral dimension has a political edge which goes against the agencies’ independent expertise, this dimension is not perceived as devoted attention to by the agencies. Whereas prior research might not provide a sufficient foundation to formulate a strong hypothesis on the substantive focus pursued by MEPs in posing their written oversight question, the rhetoric put forward by Busuioc and Rimkuté (2019) could create the basis for an expectation for an exploratory research. European agencies have been given a place in the EU regulatory state, predominantly because of their expertise. In consequence, it might be anticipated that the EP as legislative accountholder, expects the European agencies to be eager in fulfilling the technical dimension of their organisational reputation, in order to receive the EP’s support. In addition to this technical edge, the EP has shown explicit attention towards agencies’ performance (for instance via making reserves dependent on performance) (Busuioc & Rimkuté, 2019; Busuioc, 2013a). Based on this understanding and the possible clues given by Busuioc and Rimkuté (2019), this thesis anticipates that the MEPs will pursue a principal substantive focus the technical dimension and relatedly on the performative dimension, when conducting their parliamentary oversight. Since the moral dimension stand at odd with the expertise of the European agencies, this thesis will anticipate that the moral dimension will be given least attention to by the MEPs.
Expectation I (quantitative content analysis)
The parliamentary attention in terms of written posed questions is being triggered by politically salient and large European agencies
22
Expectation II (qualitative content analysis)
The MEPs will pursue a principal focus on the agencies’ technical reputation and relatedly on the performative reputation when posing their written oversight questions towards European agencies. The moral dimension of the agencies’ reputation will be given least attention to by the MEPs
23
3. Research Methodology
3.1 Research method: a quantitative and qualitative content analysis
This thesis applies a mixed research method; a content analysis carried out both in a quantitative as qualitative way (Small, 2011; White & Marsh, 2006). Holsti (1969) defined content analysis as “any technique for making inferences by objectively and systematically identifying specified characteristics of messages” (p. 14). Although the objectives of the quantitative and qualitative content analysis vary, they have been combined as research method (White & Marsh, 2006). Whereas quantitative content analysis is principally focused on a making “replicable and valid inferences from texts (…) to the contexts of their use” (Krippendorff, 2004, p. 19), qualitative content analysis purposes “to capture the meanings, emphasis, and themes of messages and to understand the organization and process of how they are presented” (Altheide, 1996, p. 33). The mixed method followed in this thesis at hand is nested, which implies that the sample studied in the quantitative content analysis is (partially) the same as the cases examined in the qualitative content analysis (Toshkov, 2016; Lieberman, 2005). This thesis uses a combined research method as the insights stemming from the quantitative research complement the knowledge generated by the qualitative research, that is to capture both the EP’s attention as the meaning (substantive focus) behind it (Lieberman, 2005). In addition, the results from the quantitative content analysis guide the case selection for the qualitative content analysis. The unit of analysis is therefore in the quantitative and qualitative thesis section the same (Toshkov, 2016). However, the level of analysis differs in both research methods, as it changes from “cross-case” in the quantitative content analysis to “within-case” in the qualitative content analysis (Gerring, 2007, p. 1; Toshkov, 2016).
3.2 Operationalization of the 2-stage design.
3.2.1 Quantitative content analysis: the parliamentary attention towards European agencies in posing written questions
The quantitative content analysis has the principal objective to examine which European agencies received most parliamentary attention in terms of written questions, and to see whether these agencies were indeed the most politically salient and largest in terms of staff
24
respect to the incidental 2009-2014 parliamentary oversight, hold as well for the MEPs of the 8th Legislature (2014-2019). In addition, it can be checked whether the suggested explanations for the empirical relations, as identified by Font and Peréz-Durán (2016), are as well valid for the 2014-2019 EP.
Written questions posed by MEPs can be accessed via the European Parliament Plenary database. The questions are here categorized according the date they have been submitted, the MEP(s) posing the written question, their political group(s) and the addressee of the question (i.e. the Council of the European Union or European Commission). With regard to the search term used to find specific written questions, the database did not provide the possibility to search for a concept in a specific word order. In consequence, searching on the full name of the European agencies would yield approximately 20.000 results per agency (see Annex II). Therefore, in the quantitative analysis the acronym instead of the full name of the European agency is used as search term, in order to secure that only those questions appear that refer to the European agencies. With respect to the search function in the database, it is possible to search for word(s) that appear in the title of the written question, and there exists the option to search for a concept that is visible in the title and/or text of the written question. Due to the limited results that became visible when searching for the acronym in the title only, the search option for title and/or written question is used. As regards the dates on which the written questions are forwarded, the quantitative content analysis covers the questions posed from January 2015 to December 2018. Despite the fact that the 8th Parliamentary term runs from 2014 until 2019, the months in the years 2014 and 2019 are excluded. This is principally due to the fact that these months were close to the election period, a reason (amongst others) which is as well put forward by Jensen, Proksch & Slapin (2013) in their study of written questions posed during the completed years of the 6th parliamentary term (2004-2009). In consequence to measure the number of parliamentary questions, it is coded how many times the acronym of the 35 European agencies (see Annex I) appear in title and/or text of the written questions posed by MEPs from the January 1st 2015 until the January 31th 2018.
The agencies receiving the largest number of total written questions are studied in terms
of political salience and staff number. For measuring the political salience of the European
agencies, this thesis makes use of the dataset on political salience by Font and Pérez-Durán (2016), which the authors provided via e-mail. To capture the political saliency of the European agencies, Font and Pérez-Durán (2016) analysed how many times the European agencies appeared in 24 printed international newspapers published in eight countries (see Annex III for names newspapers) during the period January 2010-December 2013. The authors examined the
25
public attention for the European agencies via making use of the Lexis-Nexis Academic database. This data on political salience for the period January 2010-December 2013 are perceived in this thesis as indicative for the political salience of EU agencies during the 8th parliamentary term of the EP (2014-2019). The underlying rhetoric for taking the data on political salience Font and Pérez-Durán (2016) as a lagged variable, is that the MEPs might direct their attention towards those European agencies that have been widely discussed in the printed media.
The data on the number of staff (see Annex V) in this thesis is derived from the annual report on EU agencies for the financial year 2017, published by the Court of Auditors in November 2018. The financial year 2017 is chosen in this thesis’ analysis of the 8th Legislature
(2014-2019), as Font and Pérez Durán (2016) used for their study of the 2009-2014 parliamentary term the Court of Auditors’ report for the financial year 2012. As EDA is not included in the Auditor’s report published in 2018, the data is retrieved via the agency’s 2017 annual report (EDA, n.d). Since for EUSC and EUISS the annual report for the year 2017 is not publicly available, the data is found in the 2018 annual report (for EUSC) or via information on the number of staff in 2019 as published on the agency’s website (for EUISS) (EUSC, n.d.; “About Us”, n.d.).
3.2.2 Qualitative content analysis: the reputational focus MEPs pursue in posing written oversight questions towards European agencies
As explained previously, the results of the quantitative content analysis will determine the case selection for the qualitative content analysis. The agencies receiving relatively the most written questions, are examined in the qualitative content analysis. In the qualitative content analysis mostly exploratory research is conducted with the central objective to analyse attention of the MEPs towards the European agencies by studying the substantive focus of the written
questions following organisational reputation theory.
The coding of the substantive focus following the four reputational dimensions of Carpenter (2010) –as reflective of audience expectations— proceeds in two steps. Firstly, is it checked whether the written question has the purpose to conduct parliamentary oversight of European agencies. As explained, MEPs cannot submit written questions directly to the European agencies, thus the parliamentary oversight is carried out in an indirect way. A written question is coded as a parliamentary oversight question when it focuses on the matter if/how the agency gives adherence to its mandate and/or centres on the issue of mandate itself. The
26
scrutiny in this regard can have both a positive tone (‘appraisal’) as negative (‘punitive’), following Boven’s (2004) definition of accountability. To illustrate the difference between a written question submitted by a MEP that does not have a parliamentary oversight purpose, and one that has, an example of each is given. On the 8th of May 2015, Aldo Patriciello from the PPE posed the following question to the Commission addressing cybercrime:
“Can the Commission say how it intends to go about securing greater synergy between Member State laws in this area in order to aid in its fight against cybercrime?” (Patriciello, 2015).
Patriciello (2015) referred (among others) to Europol’s figures in addressing the issue of cybercrime. However, as this question does not explicitly or implicitly address Europol’s mandate, it is not coded as a parliamentary oversight question. Broadly speaking, the questions that are not coded as having oversight purposes, are the questions that prefer clarity from the Commission or Council regarding the future strategy/steps/measures undertaken or those that would like to hear the Commission’s/Council’s view on the matter. In addition, questions are not coded as oversight questions, when they name information (for instance data or viewpoints) provided by the agency, yet do not actually address the agency any further in the question.
On 23th of April 2015, Gérard Deprez belonging to ALDE, submitted as well a written question on cybercrime. In his question he refers to Europol’s internet referral unit and ‘Check the Web’ project and asked:
“Could the withdrawal of the material not also have been placed under the remit of the ‘Check the web’ programme, which is already responsible for carrying out detection tasks? What is the justification for setting up two separate units, and what tasks are carried out by the European Cybercrime Centre (EC3 ) in this area?
What progress has Europol made in improving the circulation of information about cybercrime? Has a European platform been set up on which Member States can report offences committed on the internet? If so, what outcomes have been achieved?” (Deprez, 2015)
The question of Deprez (2015) is coded as conducting parliamentary oversight, since it is aimed at scrutinizing the operations conducted by Europol.
27
In the case a written question purposed to carry out parliamentary oversight towards the European agencies, these are qualitatively coded according to one (or more) of the four reputational dimensions of Carpenter (2010). Essential to underline is that the general tone(s) of the written question is aimed to be captured. Table 2 shows which issues are addressed by each of the four reputational dimensions.
Since a forwarded written question could include distinctive sub-questions, a question could be coded as referring to one or more of the reputational dimensions. The question could therefore be counted multiple times, with four as a maximum. To illustrate, on the 22nd of May 2018, Helga Stevens, belonging to the ECR, focused on the support provided by Frontex and asked:
“1. Does the Commission believe, also in light of the agency’s increased budget and its strengthened mandate regarding return, that Frontex can play a leading role in enhancing such checks and information‐sharing?
Performative dimension of reputation - Is the agency proficient in fulfilling their mandate? - Is the agency capable to realize their formulated
objectives in an efficient manner?
- Is the decision-making of the agency of preferred quality?
- Can the agency impose vigorous actions in striving to reach their objectives
Moral dimension of reputation - Does the agency operate in a transparent way? - Does the agency serve the common interest? - Does the agency adhere to ethical and moral
standards?
Technical dimension of reputation - Is the agency an expert in its field?
- Does the agency possess al the technical resources and knowledge needed to undertake their actions? - Has the agency the required analytical
competences?
- Is the agency scientifically and methodologically right?
Legal-procedural dimension of reputation
- Does the agency adhere to (officially laid down) procedures when carrying out their
(internal/external) actions and when adopting decisions?
- Were the procedures itself legally correct?
Table 2. The four dimensions of organisational reputation theory (Carpenter, 2010; Carpenter & Krause, 2011) applied to the written oversight questions
28
2. If so, what legal basis allows for such an operation?
3. If not, which other European agencies or initiatives would the Commission deem better suited to support Member States and on what legal basis?” (Stevens, 2018).
As Stevens (2018) addresses Frontex’ budget, mandate, role and the legal basis, this written question is coded as meeting both the performative as legal-procedural dimensions of reputation. The question posed by Stevens (2018) shows as well that not all the sub-questions can be termed as parliamentary oversight questions (i.e. sub-question 3 not). However, in the case one of the sub-questions has a parliamentary oversight purpose, the written question is coded as such and the corresponding reputational dimension of the sub-question is noted. An example of a written question coded as meeting the technical dimension is the question posed by Cornelia Ernst (GUE/NGL). On the 19th of October 2015, she asked the following question about Europol’s internet referral unit:
“What technical methods will be used to analyse open-source material and computerise online propaganda detection, and what programs developed by what manufacturers have been or will be trialled?” (Ernst, 2015).
An example of a written question coded as reflecting the moral dimension is the question forwarded by Marine Le Pen (NI) and Nicolas Bay (NI) who scrutinized the statements made by the Deputy Executive Director of Frontex;
“Does the Commission think it proper that a Frontex official is making political proposals rather than focusing on the operational management of external borders? Has the Commission responded to the comments? If so, what was its response?” (Le Pen & Bay, 2015).
It can occur that one (sub-)question includes multiple agencies. On the 3rd of July 2015, Inês Cristina Zuber (GUE/NGL) asked for instance “wat role is being played by elements of Frontex and Europol in the migrant-reception centres?” (Zuber, 2015). In the situation that the question addresses multiple agencies that are part of the case selection, the question is coded for each agency. Since MEPs are allowed to ask different institutions a written question, it might occur that the same (sub-)question is posed twice, yet to a different EU body. On the 11th of October 2018, Kostadinka Kuneva (GUE/NGL) asked the sub-question “how effective is Frontex in protecting migrants from abuse by smugglers?” to the Council, a sub-question which she
29
repeated again one month later to the Commission (Kuneva, 2018a; 2018b). If this occurs, the question is coded for each EU institution.
After the coding is completed, it is analysed which reputational dimension was most present in the total number of written questions and which dimensions dominated per agency. In addition, it is studied how the reputational dimensions changed (if so) over time. The results are reflected upon the formulated expectation, that the EP might focus predominantly on the technical reputation dimension (hereby not on the moral dimension) and on the performative
reputational dimension relatedly.
3.3 Validity and reliability of the mixed content analysis
The validity of a research design alludes to “the extent to which a measuring procedure represents the intended – and only the intended – concept” (Neuendorf, 2002, p. 122). Validity can be differentiated in internal and external validity. In securing internal validity, the researcher asks the question “are we [I] measuring what we set out to measure?” (Neuendorf, 2002, p. 125). The researcher hereby aims to bridge the gap between conceptualization and operationalization as much as possible. In the quantitative content analysis internal validity is aimed to be achieved, by using the acronym (instead of the agencies’ full name) in title and/or text, and to focus only on those acronyms that refer to the concepts intended to be analysed. In the qualitative content analysis, the internal validity is looked after by determining the coding scheme in advance and to use categories (the four reputational dimensions) which are generally speaking “exhaustive and mutually exclusive” (Neuendorf, 2002, p. 40).
External validity centres on the question if “the results of a measure can be extrapolated to other settings” (Neuendorf, 2002, p. 125). In the quantitative content analysis, the sample exists of all the 35 European agencies, being part of the wider population of independent regulatory agencies (Tohskov, 2016). The results of the quantitative analysis might be generalized to this population. In the qualitative content analysis, the cases are selected purposefully and the results can only be generalized to European agencies receiving large parliamentary attention in terms of written questions. The reliability of a research design refers to “the extent to which a measuring procedure yields the same results on repeated trials” (Neuendorf, 2002, p. 112). The quantitative content analysis can be more easily replicated than the qualitative content analysis. However, the qualitative content analysis is purposed to be made reliable by making steps and decisions in the coding as transparent as possible, and by using a coding scheme that has been applied by other academics to European agencies as well (i.e. Busuioc & Rimkuté. 2019). The reliability of the coding could have been improved by
30
intercoder reliability, in which researches independently code the same text in order to see if they come to the same results. However, in light of the nature of this research (a master thesis), securing intercoder reliability lays unfortunately outside the scope of the researcher.
3.4 Method of data collection
The principal source of evidence in this thesis at hand are the written questions posed by the MEPs. In addition, this thesis makes use of academic, policy and legal documents. Finally, data is retrieved from the website and publications of the agencies themselves.
