The Threat of a Crisis: Frontex' Reputation Management during the European Migration Crisis

(1)

Master Thesis

The Threat of a Crisis

Frontex’ Reputation Management during the European Migration Crisis

Heleen van der Donck

s2094295

Heleen van der Donck

h.j.van.der.donck@umail.leidenuniv.nl 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

International and European Governance 10 / 08 / 2018 Second Reader. | Dr. E. Bondarouk

(2)

H. vd Donck - s2093294

I. Foreword

This is it! My thesis on how Frontex deals with reputational threats during the European migration crisis. This is the end product of my MSc degree of Public Administration, and part of the specialization International and European Governance.

I’d like to thank my supervisors, Madaline Busuioc and Dovile Rimkute, for their valuable feedback. A thank you to my capstone partners, Mark Heemskerk and Floran van Amelsfort, for struggling with me, answering my phone calls, and providing a great place for discussion. A great thank you to my parents, Diana Michel and Robert-Jan van der Donck, and my brother, Jeroen van der Donck, for their endless support and love, and for offering a quiet workspace for me to write this thing. May this not scathe my reputation!

(3)

II. Table of Contents

I. Foreword 1

II. Table of Contents 2

III. List of Abbreviations 4

IV. List of Tables & Figures 5

1. Introduction 6

2. Literature Review 11

2.1 Bureaucratic reputation theory 11

2.2 Audiences 11 2.3 Reputational Threats 12 2.4 Reputational Dimensions 13 2.5 Reputation Management 14 2.6 Conclusion 17 3. Theoretical Framework 19 3.1 Dimensions of Reputation 19

3.1.1 The Performative Dimension 19

3.1.2 The Moral Dimension 20

3.1.3 The Procedural Dimension 20

3.1.4 The Technical Dimension 21

3.2 Reputational Threats & Media 21

3.3 Managing Reputational Threats 22

4. Methodology 27

2.1 Operationalizing the dependent variable: reputation management 28

2.2 Operationalizing the independent variable: reputational threats 29

2.2.1 Data Collection 30

2.2.2 Coding 31

2.3 Difference in operationalization dependent and independent variables 32

2.4 Threats to inference 33

5. Frontex, an overview 35

5.1 The Origin of Frontex 35

5.2 Mandate and Weaknesses 36

(4)

6. Results and Empirical Analysis 39

6.1 Independent variable (Reputational Threats) 39

6.1.1 Results 2013 39

6.1.2 Results 2014 41

6.1.3 Results 2015 43

6.1.4 Results 2016 46

6.1.5 Overall trends 48

6.2 Dependent variable (Reputation Management) 51

6.2.1 Results 2013 52

6.2.2 Results 2014 52

6.2.3 Results 2015 53

6.2.4 Results 2016 54

6.2.5 Overall Trends 54

7. Discussion & Conclusion 58

8. Bibliography 62

(5)

III. List of Abbreviations

AAR Annual Activity Report

AD Algemeen Dagblad

EU European Union

EUROSUR European Border Surveillance System

FDA Food and Drug Administration

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

FD Financieele Dagblad

GE Germany

IBM Integrated Border Management

JO Joint Operation

NL The Netherlands

RABIT Rapid Border Intervention Team

(6)

IV. List of Tables & Figures

Tables

Table 1 Keywords - Busuioc & Rimkute, 2018 29

Table 2 Scoring Results 2013 40

Table 6 Count Distribution - sorted per year 55

Figures

Figure 1 Distribution of Reputational Threats - 2013 40

Figure 2 Distribution 2014, sorted per country 42

Figure 3 Distribution of Reputational threats - 2014 43

Figure 4 Distribution of all articles in 2014, as counted 43

Figure 6 Distribution 2016, sorted per country 47

Figure 8 Development of the Intensity of Media Coverage & Reputational Threats 48

Figure 9 Total number of negatively coded articles, sorted by year 49

Figure 10 Total distribution of counted categories 50

Figure 11 Total Distribution of Counted Reputational Threats 50

Figure 12 Distribution of Reputation Management in 2013 52

Figure 16 Distribution of Dimensions of Reputation per Year - in percentages 55

Figure 17 Development of the Procedural Reputation 56

Figure 18 Distribution of Dimensions of Reputation per year - count 57

(7)

1. Introduction

In 2015, 1.83 million irregular border crossings were registered at the external borders of the European Union (EU), and roughly 1.25 million people applied for asylum (“Migration and Asylum”, 2018). Over 1 million people arrived by sea, of which 3,771 people drowned in the Mediterranean (“Data Portal Mediterranean Situation”, 2018).

Frontex, the European Border and Coastguard Agency, was heavily scrutinized for its lack of action in this period (The Economist, 2015). This kind of scrutiny can have serious consequences for the reputation of an agency, which in turn has consequences on the agency’s ability to cooperate with other organizations or receive funds (Carpenter, 2010, p. 49).

The significant increase of refugee arrivals in Europe in 2015 led to a rise in media attention for the migration crisis, and in turn for Frontex. In 2014 only 216,000 refugees arrived in Europe, just under a quarter of the refugees that would arrive in 2015 (“Data Portal Mediterranean Situation”, 2018).

With the start of the civil war in Syria in 2011, an increasing number of refugees made it to Europe, often travelling over the mediterranean sea, going from Turkey to Greece, or from Libya to Italy. In 2013 the Italian navy launched an operation called

Mare Nostrum to tackle the humanitarian crisis that was happening in the Strait of Sicily.

The aim of the operation was to provide security and protection for the people crossing the sea, and tackling smuggler networks and human trafficking (“Mare Nostrum Operation”, 2017).

While this operation was very effective, it was also very expensive, and in 2014 Italy decided to shut it down (“Mare Nostrum Operation”, 2017). Frontex took over their mission and started Operation Triton (ANSA, 2014). _{They had to work with less funds} and in a smaller scope of operation than Italy was operating in, which led to significant backlash (Davies, 2014). In a short amount of time Frontex went from being a relatively unknown agency, to becoming a central agency in the European debate on the migration crisis. The height of the crisis in 2015 brought ample amounts of negative

(8)

attention to Frontex.

European agencies, such as Frontex, have become an important part of the EU’s architectural set up, and are the most rapidly emerging institutional entities at the EU level (Busuioc, 2012, p. 719). There are currently more than 30 agencies operation in Europe in various areas of expertise and policy. Keleman (2002) explains that the emergency of decentralized agencies, of which Frontex is also an example, has come forth out of a political compromise between the European Commission, and the Member States. The Commission wanted to expand the EU’s regulatory capacity, and thus create more power for itself and the Union as a whole. The Member States, however, wished to maintain intergovernmental control. A solution was to decentralize certain agencies, away from Brussels (Keleman, 2002, p. 99).

Not only have EU agencies increased rapidly, literature on these agencies has also grown since the first essay collection published in 1997 (Rittberger & Wonka, 2011). According to Rittberger & Wonka (2011) the literature that has developed on this topic since then, can be divided into three different thematic threads. The first thread discusses how regulatory bodies in the EU came to be, and how their design became the way it is now. This strand also discusses the dynamics of the origin, and explores alternatives to the agency format. The second body of work discusses what happens after agencies are established. It explores the consequences and the trajectories of governance with and and by EU agencies. The third and last thread encompasses the research that questions the legitimacy and accountability of EU agencies, that are seen as independent, non-majoritarian institutions (Rittberger & Wonka, 2011, p.780).

Not only has the amount EU agencies increased over recent years, the power the have has increased as well. At first, agencies would provide information or have executive power, but recently, agencies are increasingly gaining operational and decision-making powers. Sometimes agencies even seem to have a certain amount of regulatory power (Busuioc 2012, p. 719). This is also the case with Frontex, whose mandate changed a lot in recent years, and included an increase of their powers,

(9)

beyond information sharing and executing EU policy.

However, the an increase in decision-making powers for EU agencies has an effect on the democratic character of these agencies. The distance from the EU decision-making platforms might allow agencies to escape accountability and control, which would further increase the democratic deficit already present in the EU (Busuioc, 2009, p. 600). Unfortunately, the distance and disconnect from the influence of representative EU institutions is necessary for agencies to be able to operate. Because of this independence, agencies are accused of lacking democratic legitimacy (Busuioc, 2009, p. 615). This does not mean that EU agencies are not held accountable at all. This is where bureaucratic reputational theory comes in. According to this theory, different audiences hold an agency accountable. Different audiences have different opinions on the agency, which affects the agency’s reputation. Agencies want to have a good reputation, or increase it, as it affects their ability to receive funding or to collaborate with other organizations. Different audiences affect the agencies’ reputation in different manners, as some audiences are not as powerful as others. However, for an agency losing reputation means losing legitimacy and autonomy.

In this thesis I aim to dive deeper into EU agencies and reputation, using Frontex as the main case study. Most interesting to me is: how does an agency maintain their reputation in periods of criticism? The rise and decline of the European migration crisis took place in the period of 2011 - 2017, which makes this period particularly interesting to observe. This thesis aims to take a closer look at Frontex’ reputation management around the height of the European migration crisis in 2015. Looking at media coverage during the European migration crisis, it is not unlikely that Frontex was experiencing reputational threats. A change in reputation management can be expected with an organization that is put in charge of saving lives.

In this context, this thesis will try to answer the question:

Can changes be observed in Frontex’ reputation management? If so, to what extent are

(10)

By observing Frontex, a single organization, at different points in time, two years before, during and one year after the height of the crisis in 2015, the effect of reputational threats could possibly be observed. Part of reputational theory, introduced by Carpenter (2010) are the moral, performative, procedural and technical dimensions of reputation. In this thesis, I not only aim to find out if a reputational threat will lead to a change in reputation management, I also aim to show if a specific threat, that focuses on a specific dimension of reputation, leads to a change in reputation management, that is specifically focused on a dimension of reputation.

It is relevant to discuss the way agencies react to threats, because it might help prove the democratic legitimacy of EU agencies, as Busuioc (2009) indicates they are often seen as illegitimate. If a strong response to threats can be observed, this might add to the relevant research on EU agencies, and possibly be useful for the EU agencies themselves as well, as it might show how they could gain democratic legitimacy.

While there is a body of work that applies reputational theory to US regulatory agencies and national agencies (e.g., Carpenter, 2002; Gilad, 2009; Maor, 2010), EU decentralized regulatory agencies have not been extensively observed in the context of reputational theory. This thesis contributes to the research on bureaucratic reputational theory, by using Carpenter’s dimensions of reputation and applying these to both the reputational threats and the reputation management. Carpenter & Krause (2011, p.31) state that “a critical feature of moving forward in [their] research is to examine not only the various dimensions of reputation, but to also assess their persistence and variability within and across administrative contexts”. This thesis aims to discuss the dimensions of reputation, and aims to find a new administrative context, by looking at EU agencies.

This thesis covers a specific period and organization, that has not yet been researched in the context of reputational theory. While more general studies on EU decentralized agencies, including Frontex, are emerging in the academic field, research

(11)

on reputation management around a period that might indicate a clear reason for the emergence of reputational threats, has not been done yet.

This research aims to find a causal relation between reputational threats and reputation management. To try and prove this, this thesis will be structured as follows: In the following chapter, I will give an overview of the literature of bureaucratic reputation theory, reputational threats and reputation management. The third chapter will focus on the dimensions of reputation, how reputational threats are defined, and how agencies can manage the reputational threats. Based on relevant literature, four hypotheses will be presented, that aim to explain how reputational threats influence changes in Frontex’ reputation management. In the following chapter, I will discuss the methodology, specifically how my hypotheses will be tested, and how my independent variable (reputational threats) and my dependent variable (reputation management) are to be operationalized. In the third chapter, I will elaborate on Frontex, specifying its origin, the development of its mandate and the weaknesses within the organization. In the fourth chapter, the reputational threats will be identified and the results of the analysis discussed. In chapter five, Frontex’ reputational management will be analysed using WordStat. In this chapter, I will elaborate on the results and their implications. The hypotheses will be confirmed or disproven. In the last section, the results of the analysis and the possible implications of the results will be discussed, and a final answer will be given to the research question.

(12)

2. Literature Review

This literature review will discuss bureaucratic reputation theory, and several elements within that theory. Specifically ‘audiences, ‘reputational threats’ and ‘reputational dimensions’. Furthermore, I will discuss reputation management and the different aspects of institutional architecture that an agency might take advantage of to steer their reputation.

2.1 Bureaucratic reputation theory

According to Carpenter (2010, p.45) an organizational reputation is “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”. A reputation allows an agency to receive autonomy, and it offers a guard if opposition from external audiences appears (Carpenter & Krause, 2011, p. 2). It shapes the agency’s autonomy and its administrative choice (Carpenter & Krause, 2011, p.30). An agency’s reputation influences the way it manages to maintain support and project consistency and flexibility. Reputation also has influence on the way an agency is lead during difficult times (Carpenter & Krause, 2011, p.26). According to Christensen & Lodge (2018, p. 120) reputation can be actively shaped by an agency. Therefore, understanding how a reputation is formed and managed helps understand the role of an agency in its political context.

2.2 Audiences

Audiences are individuals or groups of external actors, who observe an organization and judge it based on their observations (Carpenter, 2010, p. 33).

Audiences tend to be very diverse. Audience members can be elected officials, clients, the media, policy experts, and individual citizens (Carpenter & Krause, 2011, p. 2). The presence of diverse audiences in the bureaucratic reputation theory distinguishes it from

(13)

other bureaucratic theories.

Some aspects of a reputation can positively influence one audience’s perception of the agency, and negatively influence another audience’s perception. However, each audience also tends to prioritize certain dimensions of reputation in their judgement of the agency (Carpenter & Krause, 2011, p. 27).

Audiences evaluate an agency, based on past observations and experiences with that agency. This means that an agency’s perceived output (and with that its reputation) is not a mirror image of the real output, and that the idea that an audience has of the agency can control the agency’s general image (Carpenter & Krause, 2011, p. 6). To avoid satisfying and upsetting different audiences, agencies can sometimes present themselves in an ambiguous manner. They may not commit to a specific statement or course of action, or openly favor one audience over the other (Carpenter & Krause, 2011, p. 13). According to Carpenter & Krause (2011) public managers are extremely aware of their audiences, and the fact that they monitor them. Not all audiences are being watched closely, but some relevant or influential groups are being observed closely by agencies to calculate what is expected of the agency (Carpenter & Krause, 2011, p. 5). It is impossible for an agency to observe all audiences, therefore the agency must choose which audiences are deemed most important.

2.3 Reputational Threats

Carpenter (2010) stated that one should look at the threats an organization is facing, when trying to account for their behaviour (Carpenter, 2010, p. 832). Reputational threats provide a danger to an organization's identity and reputation. Scandals, substandard performance and public opposition, among others, generate risk for the stability of an agency’s reputation (Carpenter, 2010, p.48). In their research, Gilad, Maor & Bloom (2013) showed that reputational threats, identified as external signals, can lead to modification of structures, procedures or practices in agencies. A reputation is to be kept high or improved, as it influences the democratic legitimacy and autonomy of an agency.

(14)

Reputational threats are pressures from the external audiences, that agencies hope to satisfy with they policy. Any action of the agency can invite negative judgement and in turn reputational threat. The effectiveness of a reputational threat depends on how large and loud an audience is. Reputational threats can be based on internal processes, but a bad internal process does not necessarily lead to public outrage. Reputational threats can also be based on comparisons with other similar organizations, or on media coverage (Maor, 2015, p. 14). Furthermore, reputational threats can also have dimensions, just like the reputation of an organization. The combination of the audience that threatens the organization, and the dimension of the threat, influence how an agency is to answer to it.

2.4 Reputational Dimensions

Carpenter & Krause (2011) argue that reputation has four dimensions that influence an agency’s behavior and how audiences react to said behavior: the performative dimension, the moral dimension, the procedural dimension, and the technical dimension. The performative dimension discusses how an organization performs its duties. Does the organization do its job, and is it efficient when executing their policies? The moral dimension focuses on how an institution shows compassion and if it is judged as acting in an honest and morally correct manner. The procedural dimension highlights if an agency follows the rules. This is separate from the nature of the decision an agency makes. Technical reputation shows if it is perceived to have the capacity and skill it needs to perform its duties. While it is important to have expertise in an organization to be able to execute its policy, performative and technical reputation do not rely on each other. If expertise is present, it can still underperform, and vice versa (Carpenter & Krause, 2011, p.27).

It is not possible for an agency o have a good reputation in all dimensions, or to cultivate all dimension. This is simply infeasible (Carpenter & Krause, 2012, p. 27). Therefore they need to prioritize a dimension, or multiple dimensions, that they imagine will benefit the organization the most (or harm the least).

(15)

2.5 Reputation Management

Bureaucratic reputation theory indicates that agencies are capable of protecting their own reputation. According to Christensen & Lodge “reputation management involves issues about the core mission of an agency, reflecting on the agency’s historical path, its main resources and competences, and its outputs and outcomes” (Christensen & Lodge, 2018, p. 120). Central in managing reputation is the protection of an agency’s “turf” and the establishment of protection against hostile actors (Carpenter & Krause, 2011, p.26). Because of this, reputation management has a defensive side, that aims to shuffle blame to elsewhere, and an attacking side, where there are ulterior motives behind the cultivation of reputation, that could positively influence the content of public policy (Christensen & Lodge, 2018, p. 120 - 121).

Public organization have to deal with specific challenges that private organizations do not necessarily have to deal with. Waeraas & Byrkjeflot (2012, p. 193 - 200) have identified five.

First is the politics problem. All public organizations are in some way connected to a political level. This problem indicates that agencies only have limited influence over the kind of “turf” they can occupy. Their missions are decided on a higher institutional level, they simply have to execute the policy, and only have a small margin of discretion in what way they intend to do this. Because of this, agencies often have to perform ‘unpopular’ tasks, and during political heat, they are often an easy target (Waeraas & Bykjeflot, 2012, p. 194).

Second is the consistency problem. According to Waeraas & Bykjeflot (2012, p. 195) “Organizations with a good reputation are believed to be more consistent

in terms of their values, identities, and self-presentations, compared to organizations with a weaker reputation.” However, the diverse tasks of an agency makes it difficult to develop a consistent message (Christensen & Lodge, 2018, p. 121). Public organizations might communicate different values to different audiences, and many public organizations have responsibilities that require that application of inconsistent

(16)

sets of values, making the pursuit of consistency a substantial problem.

Third is the charisma problem. Public organizations have to appeal to our identity and make us trust them. They can no longer offer us a service that we need, it needs to make us feel good (Waeraas & Bykjeflot, 2012, p. 196). In a time where most people are not satisfied with public services, public organizations are unlikely to be received well by audiences. Furthermore, many public organizations have to deal with unsolvable issues. They might be able to improve situations, but will not be able to solve the entire problem. It is highly unlikely that all audiences will universally respond positively to that.

Fourth is the uniqueness problem. While many public organizations are specialized and expertise-based, they are not often recognized as unique. Most public organizations have common characteristics, specifically as they are all political, hierarchical, and rule-oriented entities (Waeraas & Bykjeflot, 2012, p. 198). While many organizations do differ in terms of function, transparency, organizational culture, or size, many organizations are not able to communicate this well, and audiences do not usual care to inform themselves on this (Christensen & Lodge, 2018, p. 121).

Fifth is the excellence problem. Organizations are assumed to be searching for an excellent reputation. The excellent, and best, reputation is one that is better than others. However, as public organizations often share characteristics with other organizations with the same mission, and they often have to make decisions that are not popular with audiences, it is unlikely that they will exceed their peers (Waeraas & Bykjeflot, 2012, p. 199).

These five challenges highlight that it can be very difficult for public organizations, and agencies such as Frontex, to maximise and manage their reputation. So how can an agency manage their reputation? Maor (2015, p. 26) divides the key empirical insights into two streams: (1) reputation management through changes in decision-timing, decision observability & agency outputs, more generally speaking the

decision-making process_{; and (2) reputation management through the strategic use of}

communication_.

(17)

reputation management (Carpenter 2002, 2003, 2004; Gilad & Yogev, 2012; Maor, 2011, 2015; Maor & Sulitzeanu-Kenan, 2013). According to Maor (2015) public organizations have a fundamental power in the reputation game. They are able to wait (Carpenter, 2003). Carpenter’s 2003 study shows that to ensure their reputation of an organization that ensures safety for all customers, the FDA chooses to wait and take time to approve certain drugs that might be dangerous. While a dangerous drug can be recalled, the reputation will not be mended (Carpenter, 2003, p. 492). Decision observability of agency decisions and errors is introduced in Maor’s 2011 study. He presents a different way of managing reputation. He states that agencies are actively calculating the amount of media coverage there should be a regulator’s decisions. Administrators are unlikely to put the agency in the spotlight, if errors have been made or decisions are questionable (Maor, 2011, p. 558). Carpenter’s 2002 study shows that attention given to the agency and the issue would speed up the decision-making process (Carpenter, 2002). Even more specific is Maor & Sulizeanu-Kenan’s 2013 study, which notes that negative media coverage speeds up the decision-making, while positive coverage is more likely to slow down the decision-making process (Maor & Sulizeanu-Kenan, 2013).

The second stream of literature that Maor (2015) discusses is the strand of literature discussing strategic communication. Maor, Gilad & Bloom (2012) looked into agencies’ choice to react or stay silent after facing reputational threats. They demonstrated that a regulatory agency usually stays silent on issues that they already have a strong reputation in, and they stay silent on issues that lie outside their jurisdiction. They do respond to threats that are about core functional areas in which their reputation is weaker, and areas in which their reputation is still evolving (Maor, 2015, p. 29). This research proved that agencies assess the threat, and select their communication strategy based on their reputation. Gilad, Maor & Bloom (2013) later focused on how the content of allegations has an impact on communication strategies, and which type of allegation poses a higher threat, and how they are managed. They

(18)

found that are more inclined to acknowledge problems, but will shift blame in the case of underregulation, and deny the allegations in the case of over regulation.

While Maor (2015) identifies only ways agencies can manage their reputation, there are some other ways in which it can be done as well. Waeraas & Maor (2014) identify two more ways of managing reputation, which is by shadowing practices and policies pursued by agencies that possess good reputations (as shown in research by Maor 2007, 2011), and by affiliating with established players, so they can ‘borrow’ reputation and by appointing top management teams who enjoy good reputations (Petkova, 2012, p.385).

Maor’s 2007 study states that certain highly uncertain and complex policy domains benefit from the presence of a scientific ‘gold’ standard. If this is embedded in the agency’s decision-making framework, it functions as the key legitimizing device, as it is not very susceptible to political moves. Maor’s 2011 study, as already discussed before, states that regulators that operate in low-level exposure markets, will find it very difficult to develop a positive technical reputation. To create a positive reputation they can free ride off the decisions made by expert regulators, which they can come into contact with through international collaboration schemes (Maor, 2011, p. 560).

2.6 Conclusion

So in short: agencies manage their reputation in several different ways. Reputational threats can speed up decision making, as explained by Carpenter (2002) and Maor & Silutzeanu-Kenan (2013); agencies may try to actively steer away from media attention (Maor, 2011); and reputational threats may influence how agencies communicatie (Maor et al., 2012; Gilad et al., 2013). To strengthen reputation, agencies can enforce universal excellence standards and embed them in their decision-making framework (Maor, 2007), borrow reputation through affiliation (Petkova, 2012), and they can free ride off decisions made by agencies that already have a reputation (Maor, 2011).

(19)

In this thesis, I will continue to focus on reputation management, and how agencies deal with reputational threats. The research strand I will add to, is the one of strategic communication, as I will be looking at Frontex’ external communication in Annual Activity Reports. While research on reputational threats’ influence on reputation management has been done, observing the dimensions of reputation in both the threats and the reputation management through communications has not been done. Research indicates that threats can have dimensions, but I have not found any research that applies this to a case study.

Furthermore, while some literature focuses on one specific action as a response to a threat, for example the option of waiting or not waiting, as described by Carpenter (2002) and Maor & Sulitzeanu-Kenan (2013), most literature has not looked into the strategy that forms the basis of the action taken by the agency.

Moreover, my thesis will not only look into the strategy, it will also give insight in the dimensions the agencies aim to focus on. Maor, Gilad & Bloom (2013) indicated that agencies are aware of the dimensions of reputation that are strong, but they did not research what dimension would be chosen when an agency chooses to strengthen or defend their reputation.

Lastly, the literature has shown that reputational threats, specifically through media, have an influence on reputation management, and they lead to various actions taken by the agency to safeguard their reputation. However, this mostly concerned research that focused on one event, or a short period of time, where media might be involved. My thesis will discuss a longer time period, where media attention, and threats, are almost constant.

The upcoming chapter will lay out the theoretical framework on which the rest of my research will be based.

(20)

3. Theoretical Framework

In this chapter I will conceptualize the dimensions of reputation, reputational threats and how they can be managed. Based on these three concepts, I will present my hypotheses.

3.1 Dimensions of Reputation

As briefly discussed in the previous chapter, organizational reputation is a set of beliefs about an agency that relates to the agency’s capacities and intentions (Carpenter, 2010, p. 45), and there are four dimensions to reputation: the performative, moral, procedural, and technical dimension. These dimensions all cover different aspects of an agency’s behaviour.

3.1.1 The Performative Dimension

The performative dimension focuses on the audience perception of the quality of decision-making and the achievements of the agency. Do they achieve their goals and objectives? If an agency is perceived as competent and efficient, they are often seen as having a high performative reputation. A different feature of performative reputation is the ability to intimidate certain audiences. If an agency can take drastic actions, that might not be liked by some of their audiences, to pursue their goals, they are able to display their power and decisiveness, which can positively influence their performative reputation (Carpenter, 2010, p. 46). However, depending on the audience, it can also negatively influence their performative reputation. With drastic action, often multiple audiences can end up completely polarized in their perceptions of an agency.

Depending on the size and influence of these audiences, the effect can go either way. Even when there is no drastic action, if an agency is performing exactly as it should, audiences can still view the agency negatively. This raises problems when it comes to measuring performance, as different audiences hold the agency to different performative standards (Christensen & Lodge, p. 122)

(21)

3.1.2 The Moral Dimension

The moral dimension focuses on all ethical aspects of an agency. Are means and ends morally justifiable? Is the agency transparent, and does it show compassion? Flexibility, honesty, compassion and the protection of constituents’ interests, are all important parts of the moral dimension (Carpenter, 2010, p. 46). The moral reputation is a dimension that is related most to perception and least related to facts or numbers. This means that an agency can be very effective, and yet it can still be considered having a bad moral reputation (Carpenter, 2010, p. 46). Audiences that find the moral dimension of reputation most important are likely to stand completely opposite audiences that value performance and efficiency. It can be difficult for an agency to choose which audience should be satisfied, as the agency wishes to perform well, but also wishes to not offend. This means that there is a very apparent possibility that moral acceptance will never be recognized by all audiences (Christensen & Lodge, 2018, p. 122). Agencies can emphasize their moral dimension by actively promoting aspects such as the transparency, flexibility, and socially inclusive nature of the organization.

3.1.3 The Procedural Dimension

The procedural reputation focuses on the justness of the process. Are rules and conventions followed during the decision-making process, and were these procedures thorough? While decisions may be made in an ethical manner, they might not follow common standards of procedure. If an agency follows accepted procedures to make decisions, it is likely that this positively influences their procedural reputation (Carpenter, 2010, p. 47). According to Christensen & Lodge (2018, p. 122), the procedural dimension highlights the difficulty that agencies face when having to balance compliance, responsiveness and flexibility. The dimension also highlights the linkages between procedures and outcomes. Once again, when this dimension is positive, the performative dimension is not necessarily positive, or vice-versa. It is entirely possible for an agency to follow all the rules and be completely ineffective, or be effective, yet not follow any of the rules.

(22)

3.1.4 The Technical Dimension

The technical dimension focuses on accuracy and capacity. The perception of expertise and analytical capacity influence this facet of reputation. It poses the audiences the question of qualification, specifically if they deem the agency as qualified to deal with complex situations (Carpenter, 2010, p. 46). Highlighted by Christensen & Lodge (2018, p. 122) is the difficulty to define what expertise is. This can be interpreted completely differently by both audiences and the agency itself. While an agency’s performance might be efficient, it is entirely possible that it is not qualified enough to provide specialized support and services, or the other way around. However, the performative and technical dimension are closely related, as I am of the opinion that the technical ability influences an agency’s ability to perform their duties.

3.2 Reputational Threats & Media

Reputational threats are threats to an organization’s identity and reputation. These can include scandals, substandard performance, and more generally, public opposition (Carpenter, 2010, p.48). Like reputation management, reputational threats can also be focused on a specific dimension. What is to be expected, is that an agency will focus on a different dimension of reputation than the dimension that poses a threat, as they would want to steer away from the negative connotations that arise. According to Gilad et al (2013, p. 2) agencies stay silent in domains where they enjoy a strong reputation, and threats are not present. Media coverage can involve both negative and positive comments, however, according to Gilad et al (2013, p. 9) it is more likely that audience pay most attention to the negative information, which in turn harms the agency’s reputation. It is necessary for this media coverage to be intense, as a small amount of news articles that are highly negative are less likely to reach a large audience (Gilad et al, 2013, p. 9). When audiences focus most on a certain type of threat, it is most likely that agencies do the same. Therefore it can be expected that agencies focus most on negative coverage, just like audiences. Media is able to voice reputational threats of different audiences and spread them to even more audiences. This is why it is

(23)

a relevant source to measure reputational threats. It is varied, allows for a range of opinions, and has a big reach, which in turn allows for the threats to increase, as different audiences can become exposed to the negative comments on an agency. This is one of the reasons why agencies are likely to want to resolve reputational threats.

Furthermore, the threats in media can negatively influence an agency’s current reputation. According to Maor (2016, p. 84) media coverage can make audiences aware of agencies’ actions and behavior, and it has the power to create biased perceptions of agencies ánd audiences. How agencies perceive their audience determines what information they generate for them and how feedback is evaluated (Maor, 2016, p. 84).

When agencies are faced with negative media coverage, they will try to manipulate emotions, so they can ensure a certain type of information about the agency will reach the audience, while other information does not. In my opinion this translates and generalizes to reputational threats and management as follows: when faced with reputational threats, an agency will manipulate the information audiences receive.

Negative external opinions and allegations from audiences that are transmitted through the media have the ability to negatively influence the reputation of an agency, as the opinions and allegations can negatively the perceptions of the agency that audiences have (Maor, 2016, p. 84). Negative (and intense) media coverage has the ability to negatively affect both audiences that already have negative opinions on the agencies, and audiences that do not. Therefore I think that negative media coverage can be established as a reputational threat.

3.3 Managing Reputational Threats

Agencies will steer focus towards a specific reputational dimension and away from a dimension that is perceived negatively and poses a threat to the agency’s reputation (Carpenter & Krause, 2011, p. 14). As I have shown that negative media coverage can be perceived as a reputational threat, it is to be expected that agencies try to halt or steer away from the content of the allegations. I think that agencies do this by managing their reputation, and specifically focusing on the dimensions of the reputation. The nature of the threat will also influence the nature of the shift in

(24)

dimensions in reputation management. This shift is quite difficult to do successfully, and can be done in several ways, as discussed in the previous chapter. This thesis will focus on the strategic use of communication.

As explained in the previous section, Maor (2016) stated that when faced with negative media coverage, agencies try to manipulate the audience by selectively providing certain information. This shows that an agency is essentially accentuating a certain dimension of their reputation, and steering away from another. As intense negative media coverage is necessary to force reputational management (Gilad et al, 2013, p. 9), it can be expected that changes in Frontex reputational management will occur. During the European migrant crisis there was intense negative media coverage on the way the EU dealt with (or failed to deal with) the problem. This is likely to include the actions of Frontex as well. The presence of intense negative media attention forces changes in an agency’s reputation management. Therefore this leads to the following hypothesis:

H1: _{Changes can be observed in Frontex’ reputation management during the period} 2013 - 2016.

Gilad et al (2013, p. 2) describes how underregulation poses a threat to an agency’s reputation, as it shows the agency’s failure to accomplish their goals. This indicates a failure of performance, and therefore threatens the performative dimension of reputation. Gilad et al (2013, p. 2) describe that overregulation in their specific case refers to the negative externalities of regulations, which indicates things such as continued suffering, which also influences an agencies reputation. This clearly indicates the moral dimension as being potentially negatively affected.

By analysing a small sample of newspaper articles from 2015, I was able to define initial relevant reputational threats to Frontex. In 2015, Frontex was attacked on having a lack of funding and being dependent on the EU and Member States

(25)

(Strupczewski & Baczynska, 2015), lacking resources (Mathiason, Parsons & Jeory, 2015), being overwhelmed and overworked, having a weak mandate and possibly acting undemocratically in the future (The Economist, 2015), failing to save people’s lives (BBC News, 2015, Kingsley & Traynor, 2015), and having a lack of authority (Bacchi, 2015). The majority of threats had consequences for Frontex’ performative reputation, as well as its moral reputation.

As both the exploratory search, and Gilad, Maor & Bloom’s (2013) research indicate, threats that fall into the performative and moral dimension are likely to occur. While allegations can appeal to a lack of expertise or quality, and thus refer to the technical dimension, or they can appeal to the absence of rule-following and procedures, these reputational threats are less likely to exist without seeing threats that fall into the performative or moral dimension first. In my opinion, performative and moral threats have a larger impact on an audience. It discusses whether an organization can do its job or if they are just. I think an organization is less likely to gain media attention, and thus have their reputation threatened, if the agency performs well and fairly. While they might not have the most expertise or follow the rules and procedures, they do, mostly, as they should. Audiences are simply less likely to notice what is wrong. If audiences notice an agency, it is more likely that they will notice the performative or moral side of the organization, and if this leads to negative association for the audience, they later may choose to focus on the procedural and technical aspects of the agency.

Thus, I expect that the most dominant reputational threats will be the ones that will threaten the agency’s performative dimension and moral dimension.

Maor & Sulitzeanu-Kenan (2013) showed that negative media coverage about the Food and Drug Administration’s (FDA) responsibilities as an agency, the FDA would take quick action, and communicate about their responsibilities. Maor & Sulitzeanu-Kenan also showed in their 2016 research that negative media coverage about an agency’s performance, and specifically their output, will lead to agencies increasing their outputs in the following year (Maor & Sulitzeanu-Kenan, 2016, p.32).

(26)

Both of these articles show examples of agencies that act after a threat, that falls into the performative dimension of reputation, to improve and minimize (further) reputational damage. I think that both these articles are examples of agencies laying focus on the performative dimension of their reputation, to protect their reputation from the reputational threats that express bad performance. Instead of steering away from the threat, they work to show that they cán perform well. Therefore I propose the following hypothesis:

H2: When the dominant reputational threat is related to Frontex’ _performative reputation, _{the agency will follow a reputational management strategy that focuses on}

performative reputation_.

When it comes to the reputational threats that do not focus on performance, what kind of responses can be expected? Gilad et al (2013) showed that public allegations that focus on the moral dimension, lead to agencies trying to shift blame. According to Christensen & Lodge (2018, p. 128) agencies try to relieve themselves of blame, by focusing on the procedural dimension of reputation. In a period of crisis, agencies are likely to emphasize their proper procedural approach, to shield themselves from the dangers of the reputational threat.

Hood (2011) described in his research the responses of public officials to political scandals. One of the responses he described, was that the organization would claim procedural correctness or state that the way the issue was handled was inevitable and that the outcome was unavoidable (Hood, 2011, p. 152 - 153). Therefore, I think it can be expected that agencies will respond to reputational threats that threaten an agencies moral dimension of their reputation, by applying focus to their procedural dimension. The following hypothesis comes from this:

(27)

H3: _{When the dominant reputational threat is related to Frontex’} _{moral reputation, the}

agency will follow a reputational management strategy that focuses on _procedural

reputation_.

In both the second and third hypotheses, I refer to the _{dominant reputational}

threat_{. It is entirely possible for an agency to experience reputational threats from}

multiple dimensions, but I think that an agency is most likely to respond to the most dominant threats. As the performative and moral dimension are most likely to be dominant in general, these have been chosen as the possible dominant threats in my case study.

According to Majone (1997, p. 163) the EU is a regulatory state where agencies are created to facilitate credibility. Agencies provide expertise, and as a non-elected, non-majoritarian actor, they can use their expertise to legitimate themselves, despite not being part of the electoral system (Majone, 1997, p. 157). The output of an agency is their legitimation. This leads to an emphasis on their output, and thus their technical reputation. The following alternative hypothesis thus follows:

H4: _{Frontex’ reputation management strategy will always focus on their technical}

(28)

4. Methodology

In order to test the hypotheses, I will undertake a most-similar case study analysis of Frontex, the European Border and Coast Guard Agency. The period of 2013 - 2016 will be observed, by looking at Frontex’ Annual Activity Reports (AAR) of 2013 up to 2016. In the AARs, the developments and achievements of the organization’s activities are discussed. With these reports, partners and media are informed about the details of the agency’s functioning, and are important and useful when reviewing reputation management strategies. These external communications can be used to possibly try and shift focus away from threatened dimensions of reputation. As the documents are public to all audiences, whether those are media, citizens, politicians, or interest groups, it gives the organization the opportunity to decide how they wish to present themselves and it allows them to show their priorities.

By observing data from 2013, 2014, 2015, and 2016 four (similar) cases are created: Frontex 1, 2, 3, 4. By observing the way Frontex presents itself, before and after the pivotal year 2015, the influence of the reputational threats on their reputation management can be analysed. I decided to start the analysis in 2013, to show a distribution of dimensions before 2015. I decided to analyze up until 2016, because the AAR 2016 was the last to be published. I chose to analyze 2016 to be able to show the impact of the pivotal year 2015 on reputation management and reputational threats.

I selected a single case because it allows for a deeper understanding of the workings of Frontex and because it can be expected that reputational threats are visible and identifiable. Furthermore, observing more organizations would not have been able due to the time it takes to collect and code data. As stated in the introduction, Frontex went from being relatively unknown on the international stage, to becoming a central agency in the European debate on the migration crisis, in 2015. Because of this, it is very likely that criticism in the media will provide serious reputational threats, which makes this an interesting case to analyse. The developments in the organization’s mandate throughout our observed time period also indicate it’s changing importance,

(29)

and with that, possibly, a changing reputation. This research will hopefully be able to show how Frontex dealt with that.

2.1 Operationalizing the dependent variable: reputation management

To operationalise the dependent variable of reputation management, a quantitative content analysis will be performed, using a method developed by Busuioc & Rimkute (2018). To measure Frontex’ reputation management in the period 2013 - 2016, I will use the Annual Activity Reports from each year. I will perform a content analysis on the documents, by coding words that link to the reputational dimensions. The keywords are based on the operationalization of research by Busuioc & Rimkute (2018). Documents are entered into WordStat, which counts the amount of times a word was used in each document. The program calculates general trends, frequency per Annual Activity Report, and percentages per year, which allows for comparison between the years and hopefully shows a change in focus on a dimension of reputation, which indicates reputation management. This method does not allow for much interpretation of the words, but it is structured, and can be reproduced relatively easily.

To capture the technical dimension, the keywords (see Table 1) are associated with technological and scientific practices. The keywords that capture the moral dimension are related to transparency, honesty, and credibility. The keywords that concern the performative dimension are related to results and efficiency. Lastly, the keywords that capture the procedural dimension, are related to rules, governance, and procedures. Some keywords end with ‘*’ so it allows for different iterations of the word to be counted, for example: ‘accura*’ will count all words such as ‘accuracy’, ‘accurate’, ‘accurately’, etc. The ‘_’ in ‘high*_standard*’ and ‘good_governance’ indicate a space and codes only the two words if grouped together.

(30)

Reputation Keywords

Technical accura*, analy*, assess*, data, evidence, expert*, knowledg*, measur*, methodolog*, profession*, qualitat*, quantitati*, reliab*, research, rigor*, robust, specialis*, scien*, technic*, valid*

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*

Performative achiev*, action*, chang*, control, deliveri*, effect*, effici*,

enforce*, impact*, improv*, outcom*, efficien*, reform*, request*, resist*, restrict*, result*

Procedural accessib*, fair*, formal*, good_governance, inclusive*, independent, legal*, open*, particip*, procedur*, process*, protocol*, representativ*, requiremen*, rule*, rule-driven, stakehold*,

(Table 1: Keywords - Busuioc & Rimkute, 2018)

2.2 Operationalizing the independent variable: reputational threats

To identify Frontex’ reputational threats, a qualitative content analysis will be performed. I will use a sample of newspaper articles from the Netherlands, the United Kingdom, and Germany. I will identify the main threats in newspaper articles that mention Frontex published in 2013 - 2016. I am of the opinion that a large sample, from several relevant countries in the EU will give a good overview of the criticisms of Frontex of most audiences, and will show the development of the reputational threats over the course of the analyzed period. Since I aim to show Frontex’ response to the reputational threats, it is relevant to know the dimensions of reputation the article fits

(31)

into. The sentiment of the threat present in the article will be measured into one of the four dimensions of reputation.

2.2.1 Data Collection

I will use LexisNexis to collect articles from Dutch newspapers and Factiva to

access the German and British newspapers. The keyword used was ‘frontex’, as this would show articles that really discussed this organization, and not other coast guard or border guard organizations or national organizations. Moreover, the search is limited to articles published between 01-01-2013 and 31-12-2016, to make sure it matches the time period of the Annual Activity Reports.

The Dutch newspapers that will be consulted are: Algemeen Dagblad, De Telegraaf, De Volkskrant, Het Financieele Dagblad, Het Parool, Metro, Nederlands Dagblad, NRC Handelsblad, NRC Next, Reformatorisch Dagblad, and Trouw. All of these, except Het Parool, are national daily newspapers. Het Parool is restricted to Amsterdam, except for the saturday issue, which is distributed nationally, however, with the spread of news via the internet, it has a wide reader base. De Telegraaf, as the largest newspaper, has a circulation of 393,537 issues per day, and the Nederlands Dagblad, is the smallest national daily newspaper, with a daily circulation of 19,594 (Nationaal Onderzoek Multimedia, 2017).

The British newspapers that will be consulted are: The Sun, Metro, Daily Mail, London Evening Standard, The Daily Mirror, The Times, Daily Star, The Daily Telegraph, The Daily Express, Financial Times, and The Guardian. All of these are the top daily newspapers in the UK, with The Sun circulating over 1.5 million copies in 2017 ("National newspaper print ABCs", 2018).

The German newspapers that will be consulted are: Die Zeit, Süddeutsche Zeitung, Frankfurter Allgemeine Zeitung, Die Welt, Handelsblatt, Der Tagesspiegel, Der Spiegel, Stern, Focus. These newspapers (and news magazines) have the largest circulation in Germany, with the Süddeutsche Zeitung spreading _{361,507 in the last} quarter (“_{Informationsgemeinschaft”, 2018).}

(32)

These three countries were chosen, as all three were actively part of the political debate surrounding the European migration crisis and Frontex. The Netherlands, Germany, and the United Kingdom were all contributors to the Frontex operation ‘Triton’. Furthermore, all three countries were top 10 destinations for refugees, with Germany being first on the list (Hackett, Connor, Stonawski & Potančoková, 2017). Therefore I think they all are likely to discuss Frontex in their national media. I also believe that these three countries are powerful enough in the EU that the threats from negative media coverage could influence Frontex.

2.2.2 Coding

The articles will be coded on mentions related to the four dimensions of reputations, to measure which threat was most present in each year analysed. There are three main categories: Neutral, when a mention passes no judgement on the organization; Positive, when a mention positively speaks of the organization; and Negative, when aspects of the organization are discussed that can negatively impact the agency’s reputation. The Negative category is subcategorized into the four dimensions of reputation: moral, performative, procedural, and technical.

The articles will be categorized chronologically, noting author, title, newspaper, and publication date. I will read each article, and decided in which category it fit, based on the context in which the keyword (Frontex) was placed, or in some cases, the entire article, if it discussed Frontex in more detail. Some articles might cover multiple dimensions, so multiple categories will be scored. This can only occur within the negative category, as a negative comment will always count as a threat, even when the rest of an article is positive or neutral. The positive category was included, because it has the ability to indicate a decrease of threats and an increase of a positive reputation. Simply coding everything that is not negative into the same category, would not help indicate the decrease of threats, as there is no distinction possible between neutral and positive, and it will be unclear how much of the ‘not-negative’ is beneficial to the agency’s reputation.

(33)

So how will I decide how to categorize an article? An article will be coded as ‘Neutral’, if there is no judgement (either positive or negative) present. For example: statistics as reported by the agency, information provided by a spokesperson of the agency, or simply a sentence that indicates the agency’s existence or mandate, without stating good or bad, all fall under the category ‘Neutral’.

A newspaper article will be coded as ‘Positive’ only if there are no negative statements about the agency present, and it discusses for example: the agency performing well, the agency saving lives, or the agency improving compared to how it used to be. Important to note here: if the comparison then mentions incapabilities or being specifically bad, it will not be coded as positive, but will be categorized among one of the dimensions in the negative category.

The ‘Negative’ category is divided into the four dimensions of reputation: moral, performative, procedural, and technical. An article will be coded ‘Negative - Moral’ if it mentions a breach of Human Rights, death, disrespect of employees of the agency, or inhumane behaviour. An article will be coded ‘Negative - Performative’ if it stresses the agency’s incapability to do its job or receive help from others,or if it stresses a lack of funds or materials. This also includes things that the agency doesn’t do, but the article states it should be doing (even if this isn’t part of their mandate). In the case of Frontex, there are certain expectations of the agency, that aren’t in their mandate, but are still considered ‘not doing their job’. Lastly, an article will be coded ‘Negative - Technical’, if the agency is accused of having a lack of expertise, no experts present, or if the author indicates a lack of trust in the data presented by the agency.

2.3 Difference in operationalization dependent and independent variables

While both variables are analysed within the framework of the dimensions of reputation, I chose to apply a quantitative analysis to the dependent variable and a qualitative analysis to the independent variable. My decision to analyse the newspaper articles qualitatively, was made because of several reasons.

(34)

Dutch and German and be certain that it covers and indicates the dimensions in the same manner it would in the English articles. While direct translations might be correct, they might not be used much, or specific term that are used more often in Dutch or German media would be left out. By analyzing qualitatively, this is avoided.

Secondly, the qualitative analysis, allows for nuance in the judgement of the sentiment of the articles. While some articles judge Frontex in a similar manner, and similar words are used, a small extra sentence could tip a comment from neutral to negative.

Thirdly, using the quantitative method means that it will score anything that is in the article, including sections that discuss comments that do not apply to Frontex, but perhaps another organization. This would mean that an article would be coded in a certain manner, but it would not concern Frontex at all. By performing the qualitative analysis, this was avoided.

2.4 Threats to inference

There are several possible threats to inference. First of all the susceptibility of comparative research to measurement error and random variability. I would argue that the data is concise and not fuzzy. Control of the alternative hypothesis, as well as the fact that the four cases all discuss the same organization, allows us to be sure that the data will help us uncover the causal mechanism. While this research does contain a small-N, it can focus deeply on this one organization, as there are very extensive and rich resources available that allow for analysing Frontex’ reputation management.

The possibility of generalizing this research is dubious, as it will describe just one case. This research is retrospective and accounts for this specific case, and cannot be generalized. However, the type of observation and causal mechanism could be observed in other decentralised agencies, and would allow for a larger body of work. The method applied to this case, could be applied to different case studies. Comparing the in-depth results of all decentralised agencies, would be constitute the entire population, but this is not possible in a master’s thesis.

(35)

Another common threat in comparative research is the difficulty of distinguishing association from causation. To solve this, intimate knowledge of each case, or even within-case analysis is needed. This threat is controlled by the fact that this research only analyses one organization at four moments in time, creating an in-depth analysis of this specific organization, making almost all factors of the cases the same.

To control for the explanation of the alternative hypothesis, it needs to be stated that the four ‘cases’ describe the same agency, so the reputational management strategies of Frontex should be the same to be able to argue that Majone’s argument is true, which in turn would indicate that Frontex is not influenced by reputational threats.

(36)

5. Frontex, an overview

In this chapter, I will first discuss the origin and development of Frontex, then I will discuss the mandate of the organization and its weaknesses, and lastly I will shortly give some insight into how the agency has been criticized in the past.

5.1 The Origin of Frontex

To strengthen cooperation in migration, asylum and security policies, the European Council on Justice and Home affairs created the External Border Practitioners Common Unit in 1999. The unit was made up of members of the Strategic Committee on Immigration, Frontiers and Asylum, as well as the leaders of national border security agencies. The Common Unit coordinated national Ad-Hoc Centres that educated on uniform border control. From this governance level, common operations could be implemented and overseen (Frontex, 2018b).

The European Council decided to take this cooperation effort further. In October 2004, the European Agency for the Management of Operational Cooperation at the External Borders of the Member States of the European Union (Frontex) was established through the Council Regulation (EC) 2007/2004. The aim of this agency was initially improving procedures and methods of the Common Unit (Frontex, 2018b). The agency was established to strengthen the EU borders, through a common management system and assistance of Member States’ border security activities

(Carrera, den Hartog, Parkin, 2013, p. 339).

In 2007, to assist Member States with excessive flows of ‘irregular migrants’, Frontex was mandated to deploy Rapid Border Intervention Teams (RABITs) (Carrera et al., 2013, p.340). In 2011, an amended version of the Frontex Regulation was adopted, which increased the agency’s power in coordinating Joint Operations and the deployment of European border guard teams under the supervision of a Coordination Officer from Frontex (Carrera et al., 2013, p. 340). In 2013, EUROSUR, a European Border Surveillance System was established, which would improve the exchange and

(37)

accessibility of information between Frontex and Member States (Frontex, 2017b). In 2016, after the need for a wider mandate for Frontex was expressed throughout Europe during the migration crisis, the Frontex Regulation was repealed, by Regulation (EU) 2016/1624, which established Frontex as the European Border and Coast Guard Agency (Frontex, 2018b). A need for a central European Border and Coast Guard is expressed and valued above coordinating and deploying national coast guards. In addition to running the Joint Return Operations for Member States, the new agency is now able to arrange voluntary and forced returns of migrants, deploy new RABITs to help Member States with issues like identification and registration (Slominski, 2018, p. 107). Regulation (EU) 2016/1624 led to a 60% increase in personnel and an budget increase of 18.8%, from €254 million in 2016 to €302 million in 2017. Both personnel and budget are expected to grow further in the upcoming years (Federal Ministry of the Interior, 2018).

5.2 Mandate and Weaknesses

Frontex, has as its mission promoting, coordinating and developing European border management, while following the rules of the EU fundamental rights charter and the concept of Integrated Border Management (IBM) (Frontex, 2018b). IBM is based on cooperation. Individual border agencies are more effective when there is intra-agency, inter-agency, and international cooperation. Border management becomes less effective, when cooperation is lacking in any of those dimensions ( _{Eastern Partnership} Integrated Border Management Capacity Building Project, 2018).

The legal basis of the agency is REGULATION (EU) 2016/1624, which states its responsibilities as follows: field deployment, risk analysis, situation monitoring, joint operations, rapid response, research, training, joint returns, information-sharing, and external relations.

Frontex arranges law enforcement officers, vessels and surveillance equipment for external border regions in need. (Frontex, 2018c, p. 2). Officers perform _screenings, debriefings, identification and fingerprinting of migrants. National authorities establish who gets international protection (Frontex, 2018b). This is one of Frontex’ weaknesses.