Master Thesis
Agencies and Crisis
FRONTEX reaction on the refugee crisis
Marc van der Ham
S1289373
Marc van der Ham
Marcjan.post@gmail.com
Supervisors: Dr. E.M. Busuioc | e.m.busuioc@fgga.leidenuniv.nl
Dr. D. Rimkutė | d.rimkute@fgga.leidenuniv.nl
Faculty of Governance and Global Affairs MSc Public Administration Economics and
Governance 11/ 06/ 2019
2
I. Foreword
With blood, sweat and tears I finally finished my thesis. Before reading I thank Miss Busuioc
and Miss Rimkute for the time they spent to supervise me. In addition, I will thank everyone
who read, gave advice or support me in any other way. For now, enjoy reading!
3
II. Table of Contents
I. Foreword
2
II. Table of Contents
3
III. List of Abbreviations
5
IV. List of Tables & Figures
6
1.0 Introduction
7
2.0 Literature Review
9
2.1 Reputational theory
9
2.2 Reputation
10
2.3 New Reputational Communication
11
2.4 Reputation changes
11
2.5 Dealing with reputational threats
14
2.6 Audiences
16
2.7 Role of the European Parliament
17
3.0 Theoretical Framework
17
3.1 Reputation Dimension
17
3.1.1 Performative Dimension
18
3.1.2 Moral Dimension
18
3.1.3 Procedural Dimension
18
3.1.4 Technical Dimension
19
3.2 Reputation threats and the role of the European Parliament
19
3.3 Hypotheses
20
4.0 Methodology
24
4.1 Methodology choices
24
4.2 Operationalization dependent variable
25
2.3 Operationalization independent variable
26
2.4 Coding
27
5.0 Empirical findings
28
5.1 Independent variable (Reputational Threats)
29
5.2 Dependent variable
34
4
5.4 Hypotheses tested
38
6.0 Conclusion and Discussion
40
7.0 Bibliography
43
8.0 Appendices
46
8.1 Question Examples
46
5
III. List of Abbreviations
EU
European Union
FRONTEX
Before 14-09-2016: European Agency for the Management of
Operational Cooperation at the External Borders of the Member States
of the European Union
Since 14-09-2016: European Border and Coast Guard Agency
6
IV. List of Tables & Figures
Tables
Table 1 Keywords -Busuioc & Rimkute, 2018
26
Table 2 Overall results independent variable
29
Table 3 Overall results independent variable in percentage
30
Table 4 Overall results in percentage controlling for secondary dimensions
30
Table 5 Overall results dependent variable in percentage
34
Figures
Figure 1 Question asked 2014-2017
29
Figure 2a Performative dimension in % variable in % and trend line.
31
Figure 2b Performative dimension with controlled variable in % and trend line.
31
Figure 3a Procedural dimension in % variable in % and trend line.
32
Figure 3b Procedural dimension with controlled variable in % and trend line.
32
Figure 4a Moral dimension in % variable in % and trend line.
33
Figure 4b Moral dimension with controlled variable in % and trend line.
33
Figure 5a Technical dimension in % variable in % and trend line.
34
Figure 5b Technical dimension with controlled variable in % and trend line.
34
Figure 6a Technical dimension development in the dependent variable with trend line. 35
Figure 6b Performative dimension development in the dependent variable with trend line. 36
Figure 6c Procedural dimension development in the dependent variable with trend line. 36
Figure 6d Moral dimension development in the dependent variable with trend line.
36
7
1.0 Introduction
September 2015, the refugee crisis that had already lasted for several years got another face
with the finding of a dead boy of nearly three years old at the beach of Turkey (Wagendorp,
2015). The boy and sixteen others tried to escape the Syrian civil war and fled to Europe. Their
flight had no success because their ship sunk between Turkey and Greece. This resulted in a
new peak of media attention for the refugee crisis and the safety of the European outside
borders.
The protection of the European outside borders is in the hands of two different
organizations: the national border guard of the country and the European agency FRONTEX.
The national border guard supplies the manpower and the supplies and FRONTEX delivers
some specialized employees, finance and advice. Moreover, the goal of FRONTEX is to protect
the European borders of the whole European Union and against the threat to Schengen countries
(due to the free movement of people in the EU, excluding United Kingdom and Ireland, Norway
and Switzerland) (FRONTEX, 2018).
The outside borders of the European Union are part of the European regulatory space.
Agencies, such as FRONTEX, are a special entity of the regulatory European space. Agencies
are entities created by the regulatory space to delegate tasks. These tasks consist of either the
full range of regulatory tasks such as “collecting and supplying technical information
(information-gathering) to rule-making (standard-setting) and supervision/ enforcement roles
such as to ensure consistent and coherent implementation of regulatory policy
(behaviour-modification)” or only one of those tasks (Busuioc & Rimkute, 2018).
While performing this task or tasks agencies are controlled by various audiences, for
example the European governmental actors such as the European Commission, the European
Parliament, national governments as well as the media and NGOs (Busuioc & Rimkute, 2018).
They can all check the agencies in their own manner, ranging from questions behind closed
questions of the European commission, to questions asked by the members of the European
parliament or (negative) media reports. In this jungle of different control mechanisms, agencies
have to deal with the threats they are confronted with.
As the drowned Syrian boy showed, some audiences become more interested in
checking agencies in response to events. For example, the media that reports a lot during and
right after a catastrophic event. Another example is politicians. In salient cases, i.e. cases with
8
a lot of media attention, politicians are more interested in auditing the agencies. So, during
crises some audiences pose a bigger threat to the agencies than during non-crisis time.
This research studies this phenomenon and the response of FRONTEX during the
refugee crisis period. The research question is: “How did FRONTEX manage reputation threats
during the crisis years 2014-2017”. This question will be studied by looking at the questions
asked by the Members of the European Parliament during the years 2014-2017.
The goal of this research is to search for a causal mechanism between the effects of
crisis on the reputation management of an agency. Therefore, FRONTEX was chosen as
agency, since they have had to deal with a crisis recently. Moreover, they were, in contrast to
the financial European agency, the only agency that already existed before the refugee crisis
started. On the one hand this unique position of FRONTEX makes the management of their
reputation unique. On the other hand, it can be an example for reputation management by
agencies in other crises because FRONTEX has had to deal with an enormous crisis and much
media and political attention.
To answer this research question, this study first presents a literature review. The review
starts with a short summary of the most important theories about interaction between agencies
and their audiences. Moreover, it explains why the reputation theory fits more to answer the
research question. The literature review is followed by the literature concepts of reputation,
reputation threats and their management. All these items are important for the research question
of how FRONTEX deals with reputational threats. The second part of the literature review is
about audiences and especially the role of the Members of the European Parliament as an
audience for FRONTEX.
Section three of the study starts with the in-depth explanation of the reputational
dimension, which is used later in the dependent and independent variables. Then, the interaction
between the Members of European Parliament (MEP) and the threat they pose to the reputation
of FRONTEX are discussed. The last part of section three describes four hypotheses on how
FRONTEX managed their reputation.
Section four presents the methodology in which the dependent and independent
variables are described and explained. In addition to, the coding, case selection and method are
justified.
Section five presents the findings of the research. It starts with the independent variable
of the year 2014 and goes than year by to year 2017. After that an overall picture is present
from every year, with trends. Than follows the dependent variable in the same fashion as the
independent variable, except for the year 2017. In the dependent variable the hypotheses are
9
tested and the findings are present. The last section, is a conclusion which concludes and has
suggestions for further research.
2.0 Literature review:
This section consists of the relevant concepts of the reputation theory. Moreover, it describes
theories on reputation threats and reputation management. It concludes with the role of the
European Union as audience.
2.1 Reputational theory
To understand the relationship between agencies and the regulatory space and audience there
are two theories to consider, the principal-agency theory and the reputational theory. The most
widely used theory in the field of agency relationship is the principal-agency theory, because
of the straightforward relationship between principals and agencies (Busuioc & Lodge, 2017).
The straightforward relationship is a result of the delegation of tasks from the Regulator (the
principal) to the agency. Hereby, the principal checks the agency before, during or after the
agency performs his tasks. If there is a problem with the task or the agency performs it another
way than the principal wants, the principal can hold the agency accountable. The reputational
theory assumes that this relationship is too elementary and argues that not only the principal or
principals can hold an agency accountable but other actors too. Moreover, the reputational
theory assumes that the principal is sometimes not even the most import audience in regards to
reputation.
The main assumption of the reputational theory is that the agency is looking for a unique
reputation of competence and to minimalize reputational damage (Busuioc & Lodge, 2017).
This is in contrast with the principal-agency theory, which focuses on information asymmetry,
differences between agencies and principals in information, and the profit the agency can make
off the asymmetry (Busuioc & Lodge, 2017). The reputational theory contests this idea, it
assumes that agencies hardly ever get the upper hand over the audience and are dependent on
it (Busuioc & Lodge, 2017). This results in the situation that agencies are willing to reduce
reputational damage and focus on their competence instead of to use information benefits. In
my opinion, the principal-agency theory’s scope is too limited to explain all the actions of an
agency. Therefore, in this research the reputational theory is used.
A second important assumption of the reputational theory is that agencies sometimes
take account even if they do not formally need to. And conversely, they sometimes do not take
10
account, although they were required to do so by rules (Busuioc & Lodge, 2017). The
principal-agency theory stipulates that principals (mostly elected entities) delegate some of their power
and tasks to the agency (an unelected entity) and therefore the agency is obligated to justify its
actions to the principal due the hierarchal order and possible democratic deficit (Busuioc &
Lodge, 2017, Curtin, 2007). The reputational theory acknowledges that it is possible to have
more than one audience who can give account, which is not necessarily the formal account
giver, and therefore an agency sometimes does and sometimes does not give account, depending
on which is best for its reputation (Carpenter, 2010; Busuioc & Lodge, 2017).
2.2 Reputation
The main concept of the reputation theory is reputation. Carpenter describes the reputation of
an organization as: “a set of symbolic beliefs about the unique or separable capacities, roles
and obligations of an organization, where these beliefs are embedded in audience networks”
(Carpenter, 2010: p.45). Reputation is important for an agency because it gives agencies
autonomy and options to defend themselves against critique from the audience (Carpenter &
Krause, 2012: p. 27). Moreover, reputation determines which choices agencies make, how they
handle and respond to critique. Moreover, the reputation of the agency can be effectively
changed by the agency itself, which the agency will do to act as good as possible (Christensen
& Lodge: 2018).
Carpenter describes four reputational dimensions that shape the reputation of an agency
and influences the behaviour of an agency (Carpenter, 2010; Carpenter & Krause, 2012). The
performative reputation; which focuses on the performance and effect of the agency. The moral
reputation; do agencies act in a morally acceptable manner? The procedural reputation; do
agencies follow the legal accepted ways? In addition, the technical reputation, are agencies seen
as the expert in the field of action? (Carpenter, 2010)
Carpenter argues that the actions of an organization have to been seen in light of one of
the four dimensions and, to be more precise, as a threat to one of them (Carpenter, 2010). A
threat to one of the four reputation dimensions is a threat to the agency on their functioning and
actions. An organization with a good reputation is more consistent in their values, identities,
and self-presentation than an organization with a bad reputation (Wæraas & Byrkjeflot, 2012).
Moreover, a success in reputation management leads easier task completion by, for instance, an
increase in resources (Christensen & Gornitzka, 2016).
11
2.3 New Reputational Communication.
The reputational goal for the agency is to strengthen their reputation in the fields they want to
increase. According to Busuioc & Rimkute (2018) the main instruments, besides real actions,
are the words the agencies use to describe their actions in their yearly activity rapports. Busuioc
& Rimkute used a unique dictionary to capture the four dimensions of reputation and the results
outline how agencies described themselves during their lifetime period (Busuioc & Rimkute,
2018). The goal of the agencies is to always keep their reputation and possible changes in their
own hands (Busuioc & Rimkute, 2018).
In their study, Busuioc & Rimkute addressed the questions ‘’how does the
communication of different aspects of organisational reputation vary over time and across EU
agencies?‘’ And, “can different regulatory roles account for differences in organisational
reputation cultivation strategies?” (Busuioc & Rimkute, 2018: p2). They found three answers.
First, that agencies tend to focus their communication on the technical dimension. Second, that
the moral dimension of reputation is the least used in all agency communications. And third,
overtime, agencies shift away from the technical reputation toward a more balanced form of
communication (Busuioc & Rimkute, 2018).
The shifts in communication from the technical towards a more general spreading over
the dimensions are explained in the literature, through an explicit focus of the European
Parliament on actions in the budget reports of the agencies (Busuioc, 2013; Busuioc & Rimkute,
2018). The importance of the European parliament as an audience in the European regulatory
space is new. With the increasing importance of the European Parliament as an agency the
importance of strategic communication also increased.
FRONTEX is not the only agency that uses strategic communication. The threats posed
to FRONTEX are also a form of strategic communication. The MEP use oral or written
questions to question one (or more) of the four reputational dimensions of FRONTEX. Those
questions are the strategic communication of the MEP according to Font & Duran (2016). The
literature on communicating reputation therefore recognizes not only the importance of global
strategic communication for the agencies, but also for the audience who poses a potential threat
to the agency.
2.4 Reputation changes
Reputation is not static and can therefore change over time, not only by strategic
communication, but also as a result of external events or performance. It can be increased by
good work or decreased by bad performance. Normally, agencies are aiming to better their
12
reputation or keep their reputation high/as high as possible. Although, there are situation where
getting or maintaining a high reputation is problematic.
Wæraas & Byrkjeflot looked at the differences between public and private organization
management and identify five problems regarding reputation management for public
organizations, such as agencies, and what can be done to keep their reputation as high as
possible. The problems they identified are: the politics problem, the consistency problem, the
charisma problem, the uniqueness problem and the excellence problem (Wæraas & Byrkjeflot,
2012: p.188).
The first problem, the politics problem, is the problem that arise for public organizations
”that all public organizations are connected in a political superordinate level” (Wæraas &
Byrkjeflot, 2012: p.193). The reason for this problem is that all public organizations are
instruments of elected bodies – in case of agencies of the European Parliament – that perform
public politics (Wæraas & Byrkjeflot, 2012: p.193). This leads to three problems of reputation
management. First, they task agencies to fill a need in the market. As a consequence, agencies
are not able to freely formulate their motive, vision and identity. Therefore, agencies have less
reputational possibilities than private organizations (Wæraas & Byrkjeflot, 2012). The second
management reputation problem is that agencies have to perform the mandate, regardless of
whether the public has a high or low perception of the job (Wæraas & Byrkjeflot, 2012). If the
perception of the task is low in the eyes of the public, this is often carried over to the agency,
which is usually not able to change this (Wæraas & Byrkjeflot, 2012; Hargrove & Glidewell,
1990). To conclude, the tasks of the agency affect the reputational management effectiveness
(Wæraas & Byrkjeflot, 2012: p.94). The third problem of reputation management is that politics
and agencies cannot easily be seen as separate of one another. Politicians have an opinion about
the agency and how it acts, and these negative or positive opinions are often transferred to the
agency (Wæraas & Byrkjeflot, 2012).
The second problem, the consistency problem, is the problem of public agencies in
building up a common identity and consistent ideal (Wæraas & Byrkjeflot, 2012). Agencies
have two reasons to face problems with building consistency. First, the responsibilities of the
agencies often lead to an inconsistent set of values (Wæraas & Byrkjeflot, 2012: p.195). An
example is FRONTEX, which is in charge of the border security of the whole EU. On the one
hand, they had operations in the East-Mediterranean Sea. On the other hand, FRONTEX had
for a long time no presence in Spanish Morocco. Thus, they were not consistent in defending
their borders everywhere. Second, is the problem of multiple identities (Wæraas & Byrkjeflot,
2012). Agencies, such as FRONTEX had more than one type of task and expert. For instance,
13
the task to defend the outside borders, but also to bring asylum seekers back to the country of
origin. As result, it is hard to determine one identity of an agency, or as Christensen et all state:
“all public organizations are inconsistent “by default”” (Wæraas & Byrkjeflot, 2012;
Christensen et all; Carpenter & Krause, 2012). Carpenter & Krause also made this point about
the identity of the agency: the more is unclear, the more it has to rely on its reputation (Wæraas
& Byrkjeflot, 2012; Carpenter & Krause, 2012).
Third, the charisma problem is the problem of public organizations’ lack of charisma.
“Organizations must appeal to our identity and make us feel good, admire, respect, and trust
them, not just offer us something that meets our functional needs” (Wæraas & Byrkjeflot, 2012:
p.92). One of the most important aspects of reputation is that people have an emotional
connection with the organizations (Wæraas & Byrkjeflot, 2012). The problem for agencies is
that they do not have any charisma. There are three arguments for this claim. First, agencies
cannot choose their own costumers nor can they differentiate between potential customers. Due
their mandate, agencies are often required to serve everyone in the same way, which make this
option impossible (Wæraas & Byrkjeflot, 2012). Second, due the focus on new public
management types the old style is discredited. This fluctuation makes it difficult for agencies
to build charisma (Wæraas & Byrkjeflot, 2012). Third, agencies are not associated with positive
news. They are often associated with a problem, which is not likely to go away, for example
FRONTEX and the refugee problem (Wæraas & Byrkjeflot, 2012; Hargrove & Glidewell,
1990).
The fourth problem, the uniqueness problem, is that public organizations are not seen
as organizations with a unique identity (Wæraas & Byrkjeflot, 2012). The problem for agencies
is that they have trouble to give themselves a unique image. Therefore, it is hard to defend them
self against critics on other agencies. Because, they are all seen as the same.
The fifth problem, the excellence problem, is the problem of public organizations to be
excellent or, more precise, to be seen as the right public organization for the job (Wæraas &
Byrkjeflot, 2012). Public organizations have difficulties with this because there are a lot of
public organizations with almost the same mission. For example, in the field of border
criminality FRONTEX is a major player, but there is also EUROJUST, INTERPOL and the
national police. So the question of the public and for the reputation of one of these agencies is
who is the expert? Moreover, if agencies need to make decisions that are not popular with the
public and audience, they will often not been seen as the expert (Wæraas & Byrkjeflot, 2012).
Besides these five problems of acquiring or maintaining a good reputation, there are
also situations that are a direct threat to an agencies reputation. This situation can be a scandal,
14
a crisis or other events with the common character that they are all highly salient. The European
Union regulatory space, so argues Busuioc, checked the European Parliament Committees by
using a ’fire alarm’ mechanism to check on salience and not on the whole performance of an
agency (Busuioc, 2013). Moreover, Jacobs argues that the European Parliament is not equally
interested in all agencies, but more in some than in others (2014). Egberg & Trondal also found
that bodies such as the Council and the European Parliament are more interested in highly
salient cases (Egberg & Trondal, 2011). Busuioc & Logde also state that audiences tend to
prioritize demanding accountability in salient cases to avoid reputational damage to themselves
(Busuioc & Lodge, 2016).
Not only audiences, but also individual politicians are interested in salient cases.
Ringquist & Worsham looked at the situation in the United States and concluded that high
salience leads to high attention of Congress (2003). Moreover, they argue that high salience
leads to less independency of the agency and to more control by Congress. In the European
Union, Koop argues, in highly salient cases Members of the European parliament are more
inclined to oversee agencies (Koop, 2014; Pollitt et al. 2005).
2.5 Dealing with reputational threats
As seen above, there are a several reputational threats which can occur for agencies. The next
question which in the literature arise is how to deal with those threats and even more important
which threats are the biggest problem. So, according to the literature, highly salient cases are
an important threat for reputation and have to be addressed by the agency. Christensen &
Gornitzka argue that there are two ways to deal with reputational threats and three driving forces
underneath these two ways (2016). First, the three driving forces will be discussed and then the
two ways of dealing with threats. As a supplement to Christensen & Gornitzka, Maors
distinction between reputational management will be included.
The first driving force is the rational process, which is a process whereby the leaders of
the public organization use systematic strategies to position their organization vis-à-vis
audience (Christensen & Gornitzka, 2016; Maor and Wæraas, 2015, 5). The rational process of
this method is the use of systematic strategies and a focus on economic or rational bounded
perspective topics (Christensen & Gornitzka, 2016).
The second driving force of reputational management is the image projected, which
mainly focuses on the traditional cultural values and norms of the organization (Christensen &
Gornitzka, 2016). In this process, the emphasis is on agencies’ traditions and it reflects mainly
15
the path-dependency. Moreover, the reputational tactic lies in the cultural path or the
micro-institutional context of the agency (Christensen & Gornitzka, 2016).
The third force is reputation related to wider global or national cultural context
(Christensen & Gornitzka, 2016). This includes other organizations or audiences that provide
‘object’ information where the agency focuses on. As a result agencies focus more on symbols
and how they are viewed by third parties than about the reality (Christensen & Gornitzka, 2016).
Organization reputation management can, according to Christensen & Gornitzka, be
done in two ways: developing a narrow, integrated and specific profile or a broad identifying
profile, directed at a range of audiences (Christensen & Gornitzka, 2016). The goal of the
narrow way and its advantage is that the audience knows exactly where the agency stands for.
The disadvantage is that this method does not reflect the complete overview and the complexity
of the organization (Christensen & Gornitzka, 2016). This can lead to conflicts, for example
when FRONTEX employees have to represent the brand. The broader way has the advantage
that it can communicate different messages to different kinds of audiences. The disadvantage,
however, is that trying to please every audience or sometimes trying to maintain a high
reputation with every audience leads to a situation where it is not clear where the agency stands
for, which can result in a bad reputation (Christensen & Gornitzka, 2016).
Maor made a different distinction than Christensen & Gornitzka about reputational
management. Maor argues that there are two streams: ‘’1) Reputation management through
changes in decision timing, decision observability and agency outputs and 2) reputation
management through the strategic use of communication. (Maor, 2015: p.26).
Reputational management through changes in decision timing, decision observability
and agency is based on the power agencies have for decision-making strategies with regard to
their reputation. An example of such decision-making is the power of wait (Carpenter, 2004).
Carpenter et all argue about the power agencies have over time: “Agency time discretion is the
abundant leeway that agencies have over when to make a decision, regardless of what that
decision will be”(2012: p.2). Carpenter finds in the research about pharmacy and drug agencies
that regarding decision-making options, agencies cannot overcome a bad product (2002).
Therefore, their choice for approving a drug is reputationally irreversible (Carpenter, 2002:
p.492).
Reputational management by using strategic communication is the choice of the agency
to react or to remain silent regarding reputational threats (Maor, 2015; Gilad, Maor & Bloom,
2012). Maor, Gilad and Bloom conclude that agencies often remained silent in dimensions
16
where they already have a strong reputation (2012). Moreover, they argue that agencies remain
silent in situations where they have no legal jurisdiction (2012). Besides that, agencies respond
to threats in areas where they have a weak reputation or an evolving reputation (Maor, 2015:
p.29). So, the communication of the agencies is based on the reputational threat and the status
of the agencies’ reputation.
2.6 Audiences
Audiences are, according to Carpenter, individuals or groups who observe organizations (in this
paper FRONTEX) on a regular basis, and judge them on their performance or on other aspects
(Carpenter, 2010: p.33). Audiences can be various types of persons or organizations. For
example firms, scientific or professional organizations, influential individuals or legislatures
(Carpenter 2010, Carpenter & Krausse, 2012).
Audiences can change the behaviour of agencies, namely weaken/strengthen the agency
and change the behaviour of the agency towards them. Audiences can strengthen the agency by
accepting their rules, ask them for advice or use their technical knowledge. Apart from this,
audiences can also weaken the agency by challenging their actions, advise them or by not using
their definitions. A critical point of the power of the audience is that they do not necessarily see
the reality of the agency but their own truth. Moreover, not all audiences see the agency in the
same way (Carpenter, 2010). So, agencies in public are seen through the eyes of the audiences,
who can change the behaviour of the agency (Carpenter & Krausse, 2012).
The second influence the audience can have on the regulation or the behaviour of
agencies, can lead to a change of style and rhetoric of the agency toward the audiences. Public
managers are well aware of their audiences, so they monitor them and chose the right tone for
the agencies to please the audiences (Carpenter & Krausse, 2012). Two types of audience
composition have to be identified: more or less diffused. For a less diffuse audience, they use
one type of rhetoric. For a more diffuse audience, which has a varying view of the agency, the
agency can use different rhetoric for different audiences. This difference is important for the
agency, because of the tactic they use. Lastly, not all the audiences are known, so the agency
chooses to act toward their most important audiences.
17
2.7 Role of the European Parliament
In the EU, several institutions have the task to hold agencies accountable, for example the
European Parliament, the Commission, the Council and the Court of Justice. Of these
institutions, the Commission is viewed as the most eager to hold agencies accountable (Egeberg
and Trondal 2011; Egeberg et al. 2014; Font & Duran, 2016), mainly because of the recent
interest of the European Parliament in the misconduct of agencies (Busuioc 2013; Lord, 2011).
Font & Duran argue that the European Parliament is the key actor of supervision and
holding the agencies accountable. They have a few arguments for this unorthodox statement.
First, due to new treaties, the European Parliament has strengthened their powers and
introduced more Parliamentary oversight over agencies. These increases in oversight can lead
to more agency accountability (Busuioc 2013; Lord, 2011; Front & Duran, 2016). Second, the
parliamentary oversight is a key aspect of their functioning in a democratic view (StrØm, 2000;
Front & Duran).
The European Parliament has three possibilities to enhance their agency oversight. First,
in the process of founding the agency, the Parliament can use their power in the design of
agencies to strengthen the accountability mechanism and accountability relationships between
the European Parliament and the agencies (Front & Duran, 2016). Second, the European
Parliament can use oversight by using a parliamentary commission. Third, the Members of the
European Parliament can ask questions about the conduct of the agencies (Front & Duran,
2016).
3.0 Theoretical Framework
In this part, the concepts of Reputation dimension, how to deal with threats to reputational
dimensions, accountability and how to combine them will be discussed. As a result of these
concepts, four hypotheses are introduced.
3.1 Reputation Dimension
According to Carpenter, there are four different dimensions of organizational reputation:
performative, moral, procedural and technical (Carpenter, 2010: p.45-46). These four
dimensions highlight every aspect of the reputational beliefs of an agencies’ organizational
character and their capacities and intentions (Carpenter, 2010).
18
3.1.1 Performance Dimension
The performance dimension includes whether audiences believe that (1) the agency can do their
job effectively and (2) efficiently, (3) agencies are doing the right thing and (4) agencies
intimidate the audience (Carpenter, 2010). Effectively doing their job means that an agency
made progress or finished what they had to do according to the audiences. Efficiency is about
the belief of the audience that the agency does the job without enormous budget or manpower
exceedance. Doing the right thing means that the agency does the job in a way it is supposed to
do it. Last, to intimidate audiences means that the agency is willing to take drastic actions that
are not in favour of some audience, but liked by others due the act of strength and willingness
to take action (Carpenter, 2010: 45).
An agency that does its job effectively, efficiently, by doing the right thing and if needed
intimidates has a high performance reputation. However, the problem is that not all audiences
see all the facts in the same way. Intimidating can work on some audiences, but on others it will
lower the agencies’ reputation. An agency can do the right thing according to one audience, but
the wrong thing according to another. Even if the agency works fine and efficiently by most
standards it can happen that one audience sees it differently. Therefore, it is sometimes
impossible to say that an agency has an overall high performance reputation, although it can
have one with one or more audiences.
3.1.2 Moral Dimension
The moral reputation of agencies is if they act in a morally accepted way (Carpenter, 2010).
Examples are: does the agency protect the interest of the client? Does the agency behave in an
ethically accepted way? And is it transparent? Does the agency have respect for human rights?
(Carpenter, 2010: p.46) The difficulty with the moral reputation is that it is completely separate
from the performance moral. It is possible that the agency performed very well, but handles
amoral immorally and thus has a low moral reputation. The opposite can also be true; agencies
can choose to focus on their moral reputation if their audience gives a higher degree of attention
to a moral reputation (Busuioc & Logde, 2017). Responding to this, agencies are trying to show
that they are trustworthy and protectors of values, for example of human rights (Busuioc &
Logde, 2017).
3.1.3 Procedural Dimension
The procedural reputation of an agency depends on whether the agency accepts the leading
norms and laws. This differs from the moral reputation because moral factors can be outlawed
19
and immoral actions can be in line of leading norms (Carpenter, 2010). Any audience, for
example the court or any scientific organization, can be interested in the decision-making
process and if the decision-making is in line with the norms, they accept the decision, even if
the decision is possibly amoral or strange (Carpenter, 2010). However, if the agency made a
moral, effective decision but failed to use the accepted norms and laws, the reputation of the
agency can be harmed. Due a lack of a good procedure the reputation of one dimension can
spill over to another reputation dimensions to compensate, for example immoral decisions or
disputes actions (Carpenter, 2010).
3.1.4 Technical Dimension
The technical reputation of an agency is the reputation in regard to expertise and knowledge, or
in other words, the faith the audience has that the agency has the technical capacity to do their
job (Carpenter, 2010). The main difference between technical and performative reputation is
the difference between whether the audience believes the agency is the best agency for the job
and whether the agency does a good and effective job. The difference is important because
although an agency can have a great performative reputation an audience can believe that the
agency is not the leading expert or that there is another party who is better fit for the job, and
as consequence the agency does not get the job.
3.2 Reputation threats and the role of the European Parliament
Reputational threats to an agency are threats to their performative, moral, procedural or
technical reputation. Carpenter identifies several possible causes of threats to reputation, for
example: scandals, substandard performance, and more generally, public opposition (Carpenter,
2010: p.48).
All these causes can have a negative impact on one or more dimensions of reputation.
Moreover, a bad reputation on one dimension can have a spill-over effect on other reputational
dimensions (Carpenter, 2010). Gilad, Maor & Bloom argue that agencies remain silent on
dimensions where they have a strong reputation, because most challenges are not a threat to the
established reputation (Gilad, Maor & Bloom, 2013: p.452; Gilad, Maor & Ben-Nun Bloom,
2013). They also analysed that agencies will be silent if the threat does not affect their main
reputational dimension and their overall reputation (Gilad, Maor & Bloom, 2013: p.452).
Agencies cannot maintain doing nothing in three cases: (1) if threat attacks weaken
dimensions of their reputation, (2) if their evolving reputation is attacked and (3) in cases of
high salience. (Gilad, Maor & Bloom, 2013: p.452). In case of a threat, agencies have four
20
choices: to deny that there is a problem, to admit that there is a problem or to try to blame
someone else or to take responsibility (Gilad, Maor & Bloom, 2013: p.452; Hood, 2011).
The questions Members of the European Parliament ask can be seen as an attack on the
evolving reputation of FRONTEX. FRONTEX can be considered to be in the evolving state
because of the increase in manpower and budget during 2012-2018. A growth from a small to
a bigger organization is an evolution. Moreover, since 2016, FRONTEX has a new structure
and mandate, which is a clear indication that they are in search of a new reputation.
FRONTEX is not only in a state of evolving reputation but their situation can also be
seen as a state of high salience due to the impact of the refugee crisis. A definition of salience
is, for example, the high attention of the media and other stakeholders (Christensen &
Gornitzka: 2016). In case of the refugee problem the media attention was extremely high and
they called it the Refugee Crisis (Christensen & Gornitzka: 2016). A second effect of crises is
that it affects the salience on stakeholders (Christensen & Gornitzka: 2016). So, the framing
and attention from the media makes other audiences who are important for FRONTEX, such as
the Members of the European Parliament, more interested in the case which in turn increases
the salience of the case.
A third consideration for FRONTEX is that not all of their four reputational dimensions
are considered as strong. First, there are questions about the effectiveness of operation Trition,
which was, according to some, less effective than the
preceding operation “Mare Nostrum due
Italian” (Davies: 2014). It must be noted that operation Trition has less budget than operation
“Mare Nostrum”, yet the audience saw it as a problem of FRONTEX. Second, there are
questions about the moral reputation of FRONTEX. Aas & Gundhus and Pascouau &
Schumacher questioned the human aspect and the human dignity of FRONTEX’s work and
with that they questioned the moral reputation of FRONTEX (Aas & Gundhus: 2015, p.3;
Pascouau & Schumacher: 2014, p. 1). It must be noted that this is a voice from a side of the
spectrum, which does not necessary make the global opinion.
3.3 Hypotheses
In this section five hypotheses regarding reputation management will be discussed. All these
hypotheses can be viewed from the question: “How does FRONTEX manage reputational
threats during crisis years?”
To study the research question first a distinction has to been made between crisis and
non-crisis years. During the crisis years two situations were expected. First, the increase of
21
questions of MEP, because they experience a high salient situation. The fact that they observe
a high salient situation is taken as truth for this study, since that is not in the scope of the
research. The increase of questions is the result of the high salient situation and is expected.
The study tests if the number of questions increased and with that the reputational threat for
FRONTEX.
The second expectation is that the questions of MEPs leads to a high salient situation
for FRONTEX. The European parliament is one of the main, and probably the main audience
of the European regulatory space for FRONTEX. When this audience has signs of higher
pressure on a field of experience of FRONTEX, the borders, it is likely that they will put higher
pressure on FRONTEX itself. This leads to more threats of the MEP toward FRONTEX and so
a high salient situation for FRONTEX.
Besides the increase in threats and the high salient situation, a change in behaviour is
expected. Koop, for example, states that agencies will give voluntary accountability to
audiences in cases of high salience (Koop, 2014). With voluntary accountability is mend that
FRONTEX changed its reputational management towards the threat dimension to which the
questions of MEP attacks.
The combination of high salient due the refugee crisis and the expected management reputation
change leads to the hypothesis:
H1: FRONTEX changed its reputational management during the crisis years.
Besides the change in its reputational management, it is interesting in which direction
FRONTEX changed its reputation. Therefore, the four reputational dimensions of Carpenter:
“performative”, “moral”, “procedural” and “technical” will be studied. The reasoning is simple.
The questions of the MEP attack one or more of FRONTEX’s reputational dimensions. The
response of FRONTEX to these attacks would logically also fall into one of these four
reputational dimensions. Therefore, it is important to determine why some types of threats are
likely to happen and what kind of response can be expected from FRONTEX.
The first of the threats that is expected, are threats to the performative dimension of
FRONTEX. These threats are the most expected threats to the reputation of FRONTEX,
because of the background of FRONTEX. FRONTEX increased their budget and manpower
each year from 2012 to 2018. With the increase in manpower and budget, the tasks that
FRONTEX had to perform also increased, for example, to take care of the return of refugees to
their country of origin and the task to assure the human rights of the refugees. With this increase
22
in tasks and newly added tasks the responsibility of FRONTEX increased. FRONTEX had no
history of doing these new tasks. Therefore, FRONTEX was not able to tackle threats to the
performative dimension with a strong history of performative reputation. As a result,
FRONTEX reputation was evolving.
Threats to reputation can, as the literature shows, lead to two types of responses: either
to remain silent or to actively do something (Gilad, Maor & Bloom, 2013: p.452; Gilad, Maor
& Bloom, 2013). In case of an evolving reputation, the literature suggests that remaining silent
is not an option. In addition, in the case of FRONTEX due to the threats of the questions of the
MEP remaining silent is not an option. Thus FRONTEX has to act and cannot stay silent.
Gilad, Maor & Bloom, 2013, researched the topic of under-regulation. They conclude
that, in case of regulation, the agency is likely to not reach their goal (2013). This
under-regulation can be seen as a performative failure, because the agency has a problem to perform.
Moreover, because of that, this is an attack on the performative reputation of the agency. Gilad,
Maor & Bloom, 2013 also found that the agency tackles this problem by working on their
performance. In addition, it does not focus on another aspect of their reputation, but accepts the
challenge and tries to improve the performance. Therefore, to an attack of the performative
reputation of FRONTEX the response is:
H2a: When the dominant reputational threat is related to FRONTEX’s performative
reputation, the agency will follow a reputational management strategy that focuses on
performative reputation.
Besides attacks on the performative reputation, other attacks are also possible. In most cases
other kind of threats are directed at the moral dimension. The refugee crisis distinguishes itself
due to human tragedies. The media, as a result, highlights this type of news. Moreover, because
the media have a strong interest on this topic, politicians do as well. As a result, the moral
dimension is also a dimension where threats are on.
Gilad, Maor & Bloom argue that the reaction on moral threats is that an agency will try
to avoid blame (2013). This is the expected behaviour we will see from the politicians. As
argued, politicians themselves will receive moral threats from for example the media, who
blame them for the human disasters and the lack of improvement for the humans. They will
react by blaming FRONTEX for the moral disaster. Hood reaches a similar conclusion. Hood
found that politicians try to frame their actions as the only acceptable option in terms of law
and good practice (Hood, 2011, p. 152 – 153).
23
of the MEP for FRONTEX, the expectation is that FRONTEX is willing to take voluntary
account. With voluntary account is meant that FRONTEX is willing to focus on the dimension
of threat, which the MEP poses. In case of a threat to the moral dimension, this leads to a focus
on the moral dimension for FRONTEX. This has two strategic benefits for FRONTEX. First, it
increases the relation between FRONTEX and the MEP, because FRONTEX is willing to
listing to something the MEP find important. Second, by working on a dimension on which the
MEP are also attacked on, FRONTEX increases the reputation of the MEP. This leads to a
better reputation in the eyes of the MEP, because FRONTEX helps them out in a difficult
situation. Therefore, a threat on the moral dimension is followed by:
H2b: When the dominant moral threat is related to FRONTEX’s performative
reputation, the agency will follow a reputational management strategy that focuses on
moral reputation.
The third kind of threat is on the procedural dimension of FRONTEX. This kind of threat
focuses on the legitimacy of FRONTEX to defend the borders and the question of, for example,
whether countries should have the power to defend their own borders, or more precise inside
the European Union border cases (borders between two EU member states). These threats focus
on whether FRONTEX handles in a lawful manner.
Threats on the procedural dimension are very important to tackle for FRONTEX.
FRONTEX exists because they are the organization that is tasked with protecting the European
borders against any threat, for example refugees. If MEP or other audiences dispute this task,
they dispute the whole existence of FRONTEX. In other words, if FRONTEX is not acting
according to the law they have no function at all.
The same threat exists when there are questions about any potential unlawful act of
FRONTEX while protecting the borders. Questions on these acts and these threats make it very
hard for FRONTEX to perform well. This also has a spill over effect on the performative
reputation and technical reputation. This is very harmful for FRONTEX, because it can lead to
a situation where they lose the tasks or even lose their reason for existing.
As a reaction on procedural threats, the logical answer for FRONTEX is to focus on the
procedural dimension itself. If FRONTEX is attacked on the question whether they are the
lawful organisation to do a certain job, the answer is to put a focus on the fact that FRONTEX
is a lawful organisation. Moreover, they will put a focus on the fact that other organisations or
institutions are not lawful. Therefore, any threat to the procedural dimensions is countered by:
24
H2c: When the dominant moral threat is related to FRONTEX’s procedural reputation,
the agency will follow a reputational management strategy that focuses on procedural
reputation.
Lastly, attacks on the technical reputation of FRONTEX. FRONTEX had a strong technical
reputation at the start of the crisis. FRONTEX is seen by a couple of audiences such as the
European Commission and the European Parliament as the agency to tackle the refugee crisis.
This view translates itself by extra tasks for FRONTEX. So, the technical reputation of
FRONTEX is good, otherwise FRONTEX would not be the agency that should provide the
solution, and FRONTEX should not have been asked to do these new tasks.
The good reputation in the eyes of the European regulators on the technical dimension,
leads to the situation where FRONTEX can easily tackle threats toward this reputation.
Therefore, by threats to the technical reputation:
H2d: When the dominant moral threat is related to FRONTEX’s technical reputation, the
agency will follow a reputational management strategy that focuses on technical
reputation.
4.0 Methodology
To test the hypotheses, a case study of the agency FRONTEX will be used from 2014 to 2017.
There are two main sources used in this research:
A. The Annual activity reports of FRONTEX from the years 2014 to 2016;
B. The questions of the Members of the European Parliament (MEP) during 2014 to 2017.
The annual activity reports will be used as the dependent variable. The questions of the MEP,
are used as the independent variable. I will first provide an explanation of the timeframe and
justify why the annual activity reports are used as the dependent variable and the questions of
MEP as independent ones.
4.1 Methodology choices
First, the time frame is chosen because during this period the refugee crisis received the most
attention and media salience in the EU. The attention peaks at the year 2015 and goes down the
years after that. Since this study is interested in the effect of the crisis on the accountability and
the response of FRONTEX, the research has to start before 2015. Therefore, the start date for
25
this study is 2014.
The last year that is included in the study is 2017, because of the long-time effect of the
crisis on accountability and because of the available data. To measure the long-time effect of
the crisis on accountability, the study should not only include 2016 and 2017, because although
the attention for the refugee crisis decreased, it is not zero or close to zero. So, the study should
ideally include even more years than 2018, but since that is the last year available this is not
possible.
Second, the Annual activity reports of FRONTEX were chosen as the dependent
variable, because these reports reflect the answers of FRONTEX on reputational threats and
their willingness to give accountability. The reports are checked for certain words or for a
derivative of these words. So there is a direct relation between the reports of FRONTEX if they
want to give accountability to the MEP, with a focus on the same dimension or dimensions the
MEP gives the most attention to. For example, if the MEP focus on the problems of the
performative dimension, and FRONTEX is willing to give accountability to the MEP,
FRONTEX has a focus on the performative dimension in their reports.
The independent variable is chosen because the MEP possibly have an important
audience role and therefore it is important for FRONTEX to adjust their reputation according
to the questions of the parliament. The MEP have a couple of options to check the European
agencies, for example by parliamentary questions, hearings and budgetary discharges (Font &
Duran, 2016: p.1351). Here the choice is made to use parliamentary questions because of two
reasons: “questions allow individual MEPs to solicit information and signal to the executive”
(Font & Duran, 2016: p.1351; Jensen et al. 2013; Martin 2011; Proksch & Slapin 2010; Raunio
1996; Saalfeld 2000; Wiberg 1995). “Second, questions have proven to be effective instruments
to alert the Commission about improper implementation in the member states and, on
monitoring EU agencies, written questions provide MEPs with an indirect oversight tool” (Font
& Duran, 2016: p.1351; Jensen et al. 2013). The results are from the questions that are directed
to the commission, who forwarded them to the agency, and the agencies answer them (Font &
Duran, 2016; Egeberg et al. 2014).
4.2 Operationalization dependent variable
For the operationalization of the dependent variable the results of Busuioc and Rimkute about
the annual reports of FRONTEX are used (2018). The results are in table 1. Busuioc and
Rimkute came to the numbers by creating words which are mostly referred to by one of the four
26
words: “performative”, “moral”, “procedural” and “technical” reputational dimension (2018).
The number of times they used these words are shown in table 1.
Reputation Keywords
Performative achiev*, action*, chang*, control, deliveri*,
effect*, effici*,
enforce*, impact*, improv*, outcom*, efficien*, reform*, request*,
resist*, restrict*, result*
Moral appropri*, assur*, citizen*, civil*, consumer*,
ensur*, guarant*,
guard*, health, high*_standard*, human, people, precaution*,
protect*, public, respect, safe*, save, secur*, social*, societ*,
sustain*, user*, valu*, welfare*, transpar*
Procedural accessib*, fair*, formal*, good_governance,
inclusive*,
independent, legal*, open*, particip*, procedur*, process*,
protocol*, representativ*, requiremen*, rule*, rule-driven,
stakehold*,
Technical accura*, analy*, assess*, data, evidence,
expert*, knowledg*,
measur*, methodolog*, profession*, qualitat*, quantitati*, reliab*,
research, rigor*, robust, specialis*, scien*, technic*, valid*
(Table 1: Keywords - Busuioc & Rimkute, 2018)
4.3 Operationalization independent variable
To search for threats to FRONTEX’s accountability and reputation, a qualitative research was
conducted. Therefore, all the questions asked by the MEP are searched for which threat they
posed. The questions of the MEP were chosen because some literature acknowledges them as
the main audience for European agencies (Font & Duran, 2016). Moreover, they had the option
to use all the forms of accountability checks: asked information; debated information and
imposed sanctions (Font & Duran, 2016). The MEP represents a broad spectrum of opinions
since many different parties from many different member states are present. To check the
questions, the same four reputational dimensions are used as the dependent variable. Although,
in contrast to the dependent variable, an in-depth search tactic is used. The in-depth tactic is
used to get a better picture of the questions and the motive for the questions.
27
4.4 Coding
The four reputational dimensions of Carpenter code all questions of the MEP. An extra category
is added if FRONTEX is only named as a source and not as a topic of the question. To choose
to which dimension the question is posed, the whole question was read and then added to one
dimension. Sometimes, however, the question was about more than one dimension, then one
top dimension was chosen and one or more sub dimensions. This is done, because of two
reasons. First, to add the question to only one dimension did sometimes give an incomplete
picture of the questions asked by the MEP. To tackle this, more dimensions were added.
Second, through adding a top and one or more secondary dimensions, a more realistic view of
the question emerged and it added more research options to check later. An example of more
than one dimension is:
Summary expulsions and the proposed legalisation of 'hot returns' in Spain.
(Question 000085/2014) (appendix 1)
To create a mutual understanding of the coding, an explanation of each dimension whit an
example of the database is given below.
Questions are coded performative if the ability, achievement and efficiency of FRONTEX are
questioned. Indicators that were checked during the research are words like “result”, “effect”
and “action”. Those words are only indicators for the dimensions. Additionally, the question
was read to understand the background of the question. After reading the whole question, one
or more dimensions are coded, so not only on the presence of one of the keywords. An example
of a question of the performative dimension:
Illegal immigrants channelled from Bulgaria to Greece.
(Question 012272/2015). (appendix 1)
The second category of “moral” is coded if the moral actions of FRONTEX where questioned.
Indicators such as “citizen”, “safe” and “human” were checked. After that, the context of the
question was studied to see if the MEP used a moral tone in their questions. Note that this moral
tone is subjective. After reading the whole question, not only the keyword indicators are used
but the question is coded as moral or not by its meaning. An example of a typical moral
question:
28
(Question: 015890/2015). (appendix 1)
The procedural dimension is the third coded category, which is about legal objects and law
issues. Keywords that were checked for are “law”, “rule” and “formal”. In contrast to the other
dimensions, the context was less important for coding. The reason is that the keywords are a
very precise indicator that the question is about the procedural dimension. An example is:
The migration bomb.
(Question 006706/2016). (appendix 1)
The last category of the reputation category is the technical reputation dimension. This included
questions that were asked on whether FRONTEX has the expertise to do their job or if
FRONTEX was the best actor for the job. Words as “expert”, “knowledge” and “technical”
were indicators that the technical capability was in question. More, the whole question is read
and after that the question is coded technical or not. Not only the indicators are checked. A clear
example is:
Coordination of anti-terrorism and anti-drug trafficking policies
(Question 001931/2015) (appendix 1)
The last category that is coded is the category “not for FRONTEX”. As mentioned before, this
category consists of questions in which FRONTEX is mentioned but in which FRONTEX is
not the subject of the question. An example is:
Linguistic discrimination at the Spanish Consulate in Brussels.
(Question 007487/2015) (appendix 1)
5.0 Empirical findings
In this chapter, the results of the research will be discussed and the outcomes analysed. First,
the results of the independent variable are outlined; the reputational threats posed by the
questions. Second, the dependent variable, the reputation management by FRONTEX, will be
discussed. Then these results will be reviewed in line with the theory. Last, the results will be
discussed in light of the hypotheses.
29
5.1 Independent variable (Reputational Threats)
Table 2 shows the result of questions asked by the MEP form 2014 to 2017. It shows the total
number of questions asked by the MEP by year and the total of each of the four dimensions in
each year. It is important to mention that the data before 2014 is unknown, therefore it is
impossible to known how many questions are normal for FRONTEX and how many questions
are many. The first notable category of table 2 is the performative dimension, because the
performative dimension has the largest total in each of the four years. Beside the presence of
the performative dimension, table 2 shows a difference in questions asked each year.
Table 2
Performative Procedural
Moral
Technical Totaal
2014
55
19
18
17
109
2015
190
104
71
53
418
2016
106
70
68
46
290
2017
60
49
43
23
175
(Table 2: Overall results independent variable)
Figure 1 shows the trend of total amount of questions asked by MEP using the data provide in
table 2. The first thing to note is the big increase in questions form the year 2014 to 2015. By
looking further after 2015 a decrease in questions happened in 2016 and 2017 toward almost
the same amount of questions as in 2014. This leads to a clear peak of questions asked by
MEP in 2015.
Figure 1
(Figure 1: Question asked 2014-2017)
0 50 100 150 200 250 300 350 400 450 2014 2015 2016 2017
Questions asked
30
Not only the dominance of the performative dimension and the total increase of total questions
are remarkable. The increase in questions for the procedural, moral and technical dimensions
between 2014 and 2015 is noteworthy as well as. All the dimensions received more questions
in 2015 than the performative dimension got in 2014, which was in 2014 the largest category.
This suggests at least that the MEP poses a threat to all the dimensions of FRONTEX in 2015
(by assuming that the performative dimension poses a threat to FRONTEX in 2014). The same
goes for 2016, where all dimensions, except the technical dimension, reach the number of
questions of FRONTEX in 2014. However, all four dimensions see a decrease in the total
number of questions. And even in 2017, the number of questions in the procedural and moral
dimension comes near the number of the performative dimension.
In addition, to the total numbers of questions, this research is interested in the spread of
the questions over the several dimensions. Spread of questions shows which dimension the
MEP found the most important or which dimensions the MEP found more important than
others. Moreover it shows in which dimension direction their focus is going to. Table 3 presents
the percentages of questions in each dimension in each year added total to 100%. To check the
spread of questions, the study controls for questions where, besides a primary question one or
more secondary questions are present. Table 4 shows these controlled percentages of each
dimension counting up each year to 100%.
Table 3
Performative Procedural
Moral
Technical
2014
50,5%
17,4%
16,5%
15,6%
2015
45,4%
24,9%
17,0%
12,7%
2016
36,6%
24,1%
23,4%
15,9%
2017
34,3%
28,0%
24,6%
13,1%
(Table 3: Overall results independent variable in percentage)
(Table 4: Overall results in percentage controlling for secundary dimensions) Table 4
Performative Procedural Moral Technical
2014 43,9% 21,9% 20,0% 14,2%
2015 45,4% 24,7% 18,3% 11,6%
2016 37,0% 26,9% 22,0% 14,1%
