Talking Under Threat. A mixed-methods approach to the effects of reputational threat on EU agencies' reputational management strategies

(1)

Thesis MSc Public Administration International and European Governance

- Talking Under Threat –

A mixed-methods approach to the effects of reputational threat

on EU agencies’ reputational management strategies

Mark Heemskerk

Date: 11/06/2018

S1564986

Master Thesis

Capstone:

“The

Politics

of

Organizational

Reputation: The Case of EU agencies”

Prof. Dr. M. Busuioc

Dr. D. Rimkute

Leiden University, Faculty of Governance and Global

Affairs

(2)

(3)

3

Abstract

This thesis analyses the effects of the presence of different forms of reputational threat on the reputational management strategies of EU decentralized agencies in their 2016 Annual Activity Reports. It does so by relying on a mixed-methods approach that combines the newly founded quantitative content analysis methodology of Busuioc and Rimkute (2018), with a qualitative content analysis that looks at the context in which EU agencies stress the relevant dimensions of their compound reputation. In this way the thesis aims to provide an insight into the place of reputational threats in the complex process of EU agencies’ formulation of reputational management strategies towards a large multiplicity of audiences. The thesis looks at the effects of three (non-exhaustive) forms of reputational threat on the reputational management strategies of four EU agencies. The reputational threats are labelled institutional threat,

saliency and bureaucratic competition, each with their own reputational dynamics.

Reputational management strategies rely on four specific reputational dimensions, being a

performative, technical, procedural and moral dimension. The agencies are selected from two

so-called agency categories: EU regulatory agencies and EU coordinative agencies. Comparisons are made between these two categories to find the effects of institutional threat, while comparisons are made amongst agencies of each category separately to find the effects of saliency, and bureaucratic competition. The thesis finds that institutional threat causes agencies to promote a reputation based on one dominant reputational dimension, while using the other dimensions of its reputation to support that coherent identity. Saliency threat is addressed by agencies by direct additional cultivation of the attacked reputational dimension.

Bureaucratic competition is addressed by EU agencies either differentiating their reputational

cultivation over multiple dimensions compared with other agencies in their agency category or specializing in one of the dimensions.

Keywords: Bureaucratic competition, reputational threat, Reputational management strategies, EU Decentralized Agencies, European Aviation Security Agency, European Border and Coast Guard Agency, Europol, Eurojust.

(4)

4

Foreword

Hereby I proudly present my thesis ‘Talking under Threat: A mixed-methods approach to the effects of reputational threat on EU agencies’ reputational management strategies” for the MSc Public Administration – International and European Governance at the Leiden University, Campus the Hague. And what a ride it was! The thesis was a true test of my academic mastery, that has shown me all the possible facets that belong to the research process. From unexpected surprises to exciting findings, the thesis has been of great value to me, both academically and personally. I feel as if the process has thought me more than I could ever have hoped for. Of course, every thesis process has its ups and downs, but it is more than fair to say I very much enjoyed writing my thesis!

Hereby I would like to extend a special thanks to my supervisors, Prof. Dr. M. Busuioc and Dr. D. Rimkute. I enjoyed it very much to work together with you and I am very grateful for all you have thaught me.

Also, I would like to thank Daphne Koers MSc, Kristina Becker BSc, Bert Heemskerk, Tim Heemskerk, Manon Koers, and Dick Koers. Writing a thesis is a demanding process indeed, and I am sure I would have had much more trouble finishing it properly, without all of you watching out for me.

Finally, I would like to thank Naud Berkhuizen BA for his feedback and insights that helped me to put the finishing touches on this thesis.

Mark Heemskerk (11/06/2018)

(6)

6

1. Introduction

Bureaucratic reputation theory is a rising star in the academic field of Public Administration. Ever since Carpenter’s 2010 volume ‘Reputation and Power’ provided the theoretical stream with a coherent framework for the analysis of the reputational politics of political agencies such as the USA Federal Drugs Administration (FDA) or the European Union’s (EU) decentralized agencies, the coverage of the theoretical stream has grown exponentially. This has lead the theory to cover issues ranging from the cooperative relations between EU agencies and national regulatory authorities (Busuioc, 2016) to explaining the time it takes for the FDA to decide on how to regulate a specific drug (Maor, Sulitzeanu-Keenan, 2013). Bureaucratic reputation theory has send shockwaves through the field of Public Administration and is surely expected to do so more often in the near future.

One publication that is expected to propel the theoretical field of bureaucratic reputation theory ahead, is the recent publication authored by Busuioc and Rimkute (2018) which develops a quantitative method for the measurement of so-called ‘reputational management strategies’ of EU Regulatory agencies. Their article provides a methodology through which academics can asses the reputational image EU agencies attempt to promote to the outer world using a numerical measurement, which scores and categorizes large textual sources. In other words, the methodology allows researchers to look at the way that agencies attempt to portray themselves based on the language they use in their written communication. The methodology provides a plethora of new opportunities to the field as it enables academics to quantify theoretical concepts that the theoretical field could only express qualitatively so far. The possibility of combining this new quantitative method with the traditional qualitative traditions into a mixed-methods approach is exciting. This results in a method where measurability and granularity meet.

One situation where such an approach would be desirable is the assessment of the management of the reputational image of decentralized EU agencies through the recent upheavals experienced by the EU as a whole. EU decentralized agencies have repeatedly formed very willing cases for bureaucratic reputation theory to analyse (e.g. Groenleer, 2014 ; Rimkute, 2018), and as their number has grown considerably over recent years the relevance of EU decentralized agencies to the EU’s system of governance is expected to increase (Vos, 2016). EU decentralized agencies have thus become more relevant to the state of the EU over the

(7)

7 course of the last years, and the way their reputation holds in the eyes of the audiences of the EU can be expected to have an increasingly large impact on how the EU is seen and understood by the peoples of Europe. Over the last years terrorist attacks (Little, 2016 ; Meier, 2016 ; Berthelet, 2016) and the European migration crisis (Mathiason, Parsons & Jeory, 2015 ; Ekathimerini, 2016 ; Reformatorisch Dagblad, 2016) have emerged as only two of many political dossiers that imperil the reputations that EU agencies such as Frontex or Europol hold towards the wider public. Following the lines of thought of the bureaucratic reputation theory (Carpenter, 2010a), these kinds of threats to the reputation of these agencies put pressure on the agencies to alter the strategies through which they build their reputation in the eyes of their audiences. The theoretical link between these so-called ‘reputational threats’ and the self-image organizations try to establish with the larger public is one that has been repeatedly assumed by various theoretical accounts of the field of bureaucratic reputation theory (Carpenter, 2010a ; Carpenter & Krause, 2012 ; Busuioc & Rimkute, 2018). The field however still misses a comprehensive assessment of how different forms of reputational threat affect the way that EU agencies present themselves; an approach that provides an assessment that is both measurable and testifies of granularity, looking at the effects of reputational threat while taking into account the entire process of how EU agencies manage their reputations in a political arena, ruled by reputational politics. This thesis occupies itself with that exact issue. How do threats to an EU agency’s reputation affect the way in which the agency presents itself to the outer world?

1.1 Main Research Question and Relevance of the Thesis

This thesis provides a deeper insight into how decentralized agencies of the European union manage their reputation towards a multiplicity of audiences. It looks at the role of different forms of reputational threat and how these affect the process that an agency goes through in determining its reputation management strategies. It makes use of the new possibilities that the recently developed quantitative methodology provided by Busuioc and Rimkute (2018) has brought to the field, by harnessing elements of the methodology into a mixed-methods approach which this thesis relies on. Busuioc and Rimkute’s (2018) methodology uses keywords to identify reputational management strategies in any written communications. Doing so allows for a closer inspection of how agencies shape the language they employ to present and legitimize themselves towards their audiences, and it forms a useful tool for any academic that wishes to investigate reputational management strategies of any type of organization. In this light the thesis adheres to the following main question to guide its research.

(8)

8 “How does the presence of reputational threat affect the usage of reputational

management strategies by EU decentralized agencies in their annual activity reports of the year 2016?”

The thesis focuses on the rhetoric of EU decentralized agencies in their annual reports of 2016 to gain an insight into how reputational threat may affect reputational management strategies. It will apply a mixed-method approach to a small-n comparative case study to consolidate explanatory hypotheses that can form guidelines for venues of future research. In this sense the thesis drafts hypotheses, and through an explanatory fashion assesses the prominence and potential of each of these hypotheses for further research both quantitatively and qualitatively.

Through this process the thesis gains its academic relevance. First of all, the thesis aims to answer the call of Busuioc and Rimkute (2018), who in their article asked for further academic research to investigate what elements cause further variety between EU agencies in terms of their reputational management strategies. One specific line of inquiry that they suggest is the comparison of different agency types; the comparison of regulatory agencies with executive/coordinative agencies to be exact. As this thesis conceptualizes one specific form of reputational threat (Institutional threat) as closely linked with agency category the thesis works to follow this venue and assesses its potential. Secondly, the thesis contributes to our conceptualization and understanding of reputational threat. Where reputational threat has formed an important role in the academic field of bureaucratic reputation theory the concept has been sparsely defined and accordingly is used in different, limited fashions that keep other forms of reputational threat out of the scope (e.g. Maor & Sulitzeanu-Keenan, 2015 ; Grøn & Salomonsen, 2017). In contrast to the existing literature this thesis does not treat reputational threat as a unitary concept but explains how the concept holds multiple forms and natures. These forms of reputational threat are not to be understood as an exhaustive list or as mutually exclusive, but rather aim to show how different threats have different effects on an agency’s reputational management strategies. In doing so the thesis both shows the complexity of the concept and uses these multiple forms to provides a clear-cut definition that can be used in further research to indicate a reputational threat to an agency. Reputational threat is a prominent concept in the bureaucratic reputation theory literature and can be seen as a ‘usual suspect’ when looking for phenomena that affect reputational management strategies (Carpenter, 2010a ; Busuioc & Rimkute, 2018). Therefore, assessing the effect of this concept using the new

(9)

9 method of Busuioc and Rimkute (2018), it is expected this thesis will find interesting outcomes that can help deepen our understanding of the relationship between the two concepts.

Besides adding this novel line of inquiry to the findings of Busuioc and Rimkute (2018), the thesis also aims at contributing in a methodological sense, by showing how the methodology can be used as a quantitative side to the thesis’s mixed-method approach. Up until so far, most of the studies that have employed bureaucratic reputation theory have done so by means of qualitative single-n case studies (Maor, 2010 ; Carpenter, 2010a), through qualitative small-n comparative case studies (Rimkute, 2016) or through quantitative large-n studies (Busuioc & Rimkute, 2018). This thesis shows how the understanding of qualitative evidence can be enriched using quantitative descriptive methods to point the assessment of the evidence in the right direction. This thesis does so by first using the quantitative results derived from applying the methodology of Busuioc and Rimkute (2018) and then assessing the results in a qualitative analysis of the content of the same evidence. A complete oversight of how this mixed-methodology will be enacted can be found in the methodology section of the thesis.

Finally, the societal relevance of the thesis is to be found in its theoretical explanation of EU regulatory agencies’ behaviours in their communications through annual reports. Understanding how an agency speaks about its work towards its larger audiences and understanding what dynamics underlie the words the agency uses to communicate with larger society, contributes to our ability to hold these agencies accountable for their actions. In that sense, this thesis conclusions about the causal mechanisms that are fundament to the agencies’ wording of their annual report contributes to political actors’ and other academics’ ability to contextualize the annual reports of EU agencies in general and causes them to be better able to understand why the agency says what it does.

1.2 Structure of the Thesis

The thesis is structured into five sections and is furtherly guided by five sub-questions. The purpose of these sub-questions is to support and structure the thesis’s investigation into the main research question. The sub-questions together provide an insight into all the relevant elements of the main research question and outline the process this thesis goes through to provide a coherent answer. In this way the sub-questions strengthen the thesis internal validity and provide additional structure to the thesis itself.

(10)

10 The first section of the thesis is the literature review, which aims at constructing a theoretical framework that provides a coherent set of theories to be employed for the thesis’s analysis. The literature review dives into the existing literature in the field of bureaucratic reputation theory to (i) provide an oversight of the origins, assumptions, findings and general usage of bureaucratic reputation theory, (ii) to conceptualize and define the relevant concept necessary to answer the thesis’s main research question, and (iii) to provide an insight into the state of the art of bureaucratic reputation theory. To this end this section is structured around the sub-question:

SQ1: ‘What does the existing academic literature say about the relationship between

reputational threat and reputational management?’

This sub-question takes stock of the exiting literature to compile a theoretical toolbox made up of the relevant concepts and mechanisms that aid in answering the thesis’s main research question. The sub-question is the starting point for further investigation as its findings will provide a theoretical context for the assessment of the further sub-questions. Additionally, also the theoretical framework and the hypotheses guiding the research are outlined and theoretically motivated.

The second section of the thesis is dedicated to explaining and justifying its methodological choices. To this end it assesses the steps that are taken to safeguard the validity and robustness of the results of the thesis. The most essential elements discussed in this section are the operationalization of the used variables, the thesis’s research design, its exact methods and finally its case and source selection. To this end the section first exhaustively outlines how the reputational threats and reputational management strategies will be employed and measured in the later sections in the thesis. The clear and full outlining hereof strengthens the venues for replicability of the research and guides the reader through the later sections. Secondly, the section looks at how the relationship between the variables is analysed exactly, providing a step-by-step outline of how the research is conducted and what its aims are. Thirdly, the means and methods by which the research is conducted are outlined. Finally the section motivates why and how the cases and used sources that the research derives its inferences from are selected and categorized following the operationalization of the variables.

The third section of the thesis is the results section, which discusses the results from the quantitative side of the thesis’s mixed-method research design. The section is divided into two parts, each of which discusses one of the two lines of inquiry of the thesis. That is, one of the

(11)

11 two looks at the effect of institutional threat on reputational management strategies, and thus compares agencies along the lines of their agency category. This part answers the following sub-question:

SQ2:‘How does the usage of reputational management strategies differ among EU

decentralized agencies with different institutional reputational threats?’

The second part looks at reputational threat between agencies within the same agency category and thus looks at the influence of reputational threat as saliency and reputational threat as bureaucratic competition on reputational management strategies. The sub-question that guides this sub-section flows as followed:

SQ3: ‘How does the usage of reputational management strategies differ between EU

decentralized agencies, that are and that are not subject to reputational threat as saliency and reputational threat as bureaucratic competition?’

The distinction between the two sub-questions is theoretically motivated and is justified in the thesis’s methodological section. In the results section these questions are assessed using the sections’ descriptive quantitative evidence that has been compiled assessing the selected cases using the datamining and statistical software-tool WordStat.

The fourth section is the analysis section. In the analysis section inferences are drawn from the findings in the results section, which are used to support the thesis’s qualitative content analysis. This means the thesis looks at the language usage in the annual reports of the selected EU agencies and assesses the context in which the language expressing certain reputational dimensions are used. In doing so the section uncovers the causal mechanisms that underlie the relationship between the different forms of reputational threat and the various reputational management strategies of the selected EU agencies. In this way the exact relation between the variables is investigated, and evidence for answering the thesis’s main research question is provided. The two sub-questions to which this section provides an answer are:

SQ4: ‘What patterns can be found can be found holding the results of the quantitative

analysis of the agencies’ reputational management strategies against the textual context of the agencies annual activity reports?’

SQ5: ‘To what extent can the found patterns in reputational management strategies be

(12)

12 The final section of the thesis is the conclusion. In the conclusion the findings of the earlier sections are summarized and are combined in such a fashion to provide an answer to the main research question of the thesis. Hereafter there is an additional discussion section in which limitations of the research are shortly discussed and venues for further research are suggested.

(13)

13

2. Literature Review

In this section of the thesis the existing body of bureaucratic reputation literature will be outlined and assessed. It (i) provides an oversight of the main elements of bureaucratic reputation theory, (ii) brings an insight into the state of the bureaucratic reputation theory and (iii) gives an in-depth conceptualization of the independent and dependent variables of the thesis and how they interact. In this way the literature review section aims both at constructing a theoretical toolbox that can be used to exercise the analyses of the further sections and at providing an assessment of what new insights bureaucratic reputation theory has brought to the field of Public Administration. The sub-question this section addresses is ‘What does the

existing academic literature say about the relationship between reputational threat and reputational management?’. This section is divided into seven sub-sections, each of which

adds to answering the section’s sub-question. The first subsection provides an oversight of the main assumptions, ideas and theoretical viewpoints that form the core of bureaucratic reputation theory as employed in this thesis. The second sub-section provides a coincided insight into how bureaucratic reputation theory has contributed to the academic field both in terms of (i) concepts and theories and (ii) the coverage of cases that bureaucratic reputation has helped to explain. The third sub-section provides a conceptualization of the concept

reputational management strategies and distinguishes a typology widely used in bureaucratic

reputation theory literature that will be too employed in this thesis. Fourthly, this section addresses the conceptualization of the concept reputational threat. Doing so it distinguishes three forms of reputational threat that are commonly perceived in the theoretical field’s literature. The sub-section then continues to formulate a clear-cut definition of the concept of reputational threat. The fifth and sixth two sub-sections come in the form of a processing of the theoretical insights of the rest of the section and use the found theories and concepts to construct a theoretical framework and formulate six hypotheses that will guide the research to come to meaningful conclusions to this thesis’s main research question. Finally, the final seventh sub-section provides a section conclusion.

2.1 Bureaucratic Reputation Theory

The main, leading works to the theory of bureaucratic reputation as referred to in this thesis, have been written by Daniel Carpenter (2001 ; 2010a), who first emancipated the idea that the image that bureaucracies hold to relevant audiences in their political network mattered

(14)

14 for their political power. He believed that this relation of ‘bureaucratic reputation’ formed the basis for the ‘autonomy’ and the ‘power’ that these bureaucracies hold as organizations. Carpenter (2001) first introduced his take on the theoretical link between bureaucratic reputation and the political abilities of bureaucracies in his 2001 seminal work ‘The Forging of Bureaucratic Autonomy’, in which Carpenter provided explanations for the autonomy of multiple bureaucratic actors through a large variety of theoretical traditions, including rational-choice theory and organizational theory. In this volume Carpenter (2001) emancipated the idea that bureaucratic autonomy is in fact closely linked to the reputation bureaucracies carry in a moral, efficiency or expertise sense and is embedded in a highly complex context of political links to external audiences. Building on these findings, Carpenter’s 2010 work provides a much more elaborate account of this idea and to the day of today remains to be the central work in the theoretical field. Carpenter (2010a) provides an extensive account of how the politics of bureaucratic reputation have allowed the FDA to secure and expand a niche of authority within the medical world in the United States. To Carpenters belief, ‘reputation’ as a concept refers to a construct that is ‘composed of symbolic beliefs about an organization – its capacities, intentions, history, mission – and that is embedded in a network of multiple audiences’ (Carpenter, 2010, p. 33). Reputation as such is a medium of legitimacy, that places the organization in a multidimensional and dynamic political arena. Through their image in the eyes of other actors, organizations are affected by the beliefs these other actors hold about them, either to constraining or to enabling them (Carpenter, 2010). Reputation should therefore not be seen as a formalized or even as a constitutionalized characteristic of an organization’s founding documents. Rather reputation is a trait that organizations earn and carry through symbolism in their relations with other actors within their network: their so-called ‘audiences’ (Carpenter, 2010a). Audiences in this context indicate any individual or group of individuals that assesses the behaviour of an organization and that judges this behaviour along the lines of its held perceptions on how this organization should behave. Audiences include amongst others: media groups, scientific communities, the mass public and other political entities. Each of these audiences bases its vision on the reputation-holding organization through its own perceptions of the organizations authenticity, intentions, legalities, capabilities, weaknesses, history, and various other elements (Carpenter, 2010a) As an actor is embedded in a network of audiences, that do not necessarily all hold the same ideas with regards to how the organization should act, this produces an interesting dynamic between reputation-holders and their audiences. Not only do audiences grant the organization the legitimacy to act by either accepting passively or supporting actively the organization’s behaviour, but also do

(15)

15 organizations - both consciously and unconsciously - adapt their behaviour to the wishes and criticisms of their audiences. This adaptation of behaviour happens both in actual policy-behaviour as well as in the organization’s rhetoric towards its audiences (Carpenter, 2010a). As an organization’s audiences are however compound and hold different perceptions of the organization, the organization can hold multiple reputations at once over different policy areas or in the eyes of multiple audiences. This means that when the organization fails to satisfy its audiences, this feeds back into the organizations held reputation towards these audiences. This political game of legitimacy through reputation forces the organization to constantly steer and strategize over how it wishes to be perceived by what audiences (Carpenter, 2010a). This political game of legitimacy is reminiscent of Easton’s (1965) systems theory, which theorizes the relation between the regulatory system and the environment it regulates to be a system of input, throughput and output. The environment provides the political system with demands and legitimizing support, which the political system then processes and uses to finetune its policy behaviour. This behaviour is then feeds back into the demands and support of the environment of the system. When the demands on the system becomes too high, the support too low or an external shock disrupts the cycle, we speak of stress and the resilience of the political system determines how well the system will be able to survive.

The reputational dynamics of the political game of legitimacy in which bureaucracies are embedded give way to three main challenges that are guided by their relevant reputations: (i) How do organizations maintain a support-base, broad enough to legitimize their

behaviour?

(ii) How do organizations steer their way through the reputational assessments of their audiences, both of their supporters and of their enemies?

(iii) And how do organizations manage to maintain a consistent line of policy output while guaranteeing enough flexibility to react on the wishes of its audiences? (Carpenter & Krausse, 2012).

2.2 The Academic Impact of Bureaucratic Reputation Theory

Bureaucratic reputation theory has been employed in various ways and has provided the field of Public Administration with a large range of insights that, as Daniel Carpenter (2010a) himself had promised, have allowed for an insight of the psychological dynamics of public officials looking for esteem, without the theory trying to be universalist in its endeavours. The theoretical tradition started by Carpenter has found a large resonance with the

(16)

16 academic community, and bureaucratic reputation theory has given way to a plethora of cases, findings and concepts, which have supported and expanded the academic field.

Bureaucratic reputation theory has worked to redefine many traditional concepts of Public Administration theory. Concepts such as bureaucratic autonomy, accountability and legitimacy, that have a traditional Weberian origin, have been expanded through the social mechanisms of bureaucratic reputation theory. Groeneleer (2014) contended that bureaucratic autonomy, rather than having a purely constitutional basis, is in fact the product of a political game that agencies play to amass support for the legitimacy claims that motivate their autonomy. The ways in which organizations are theorized to be able to do so are for example by differentiating itself from similar agencies in the field. Through his endeavours Groeneleer (2014) managed to add an additional dimension to the concept of autonomy. Rather than being the pure product of constitutional, material or value-based factors (e.g. Bauer & Ege, 2016 ; Danielsen & Yesilkagit, 2014), Groeneleer (2014) points out the social factors of a bureaucracy’s image and its interrelations with other political factors that affect an agency’s autonomy, and expands our theoretical understanding of what fundaments a bureaucracy’s autonomy and legitimacy. Likewise, Busuioc and Lodge (2017) have reconceptualized accountability beyond its traditional understanding as a feature of bureaucratic hierarchy and instead have pointed at the reputational dynamics that underlie its interplay. In this vision accountability is a relation between two types of actors: account-givers and account-holders. Each of these actors carries its own reputational concerns for effectively pointing out the flaws in the other’s behaviour. In this sense the ability to live up to one’s responsibility (keeping the other actor in account) becomes a determinant of one’s own reputation (Busuioc and Lodge, 2016). It is in this sense that Busuioc and Lodge (2016 ; 2017) point out the core-differences between bureaucratic reputation theory and principal-agent (P-A) theory. Traditional accounts based on P-A theory (e.g. Kassim & Menon, 2003 ; Miller, 2005) tend to conceptualize relations between agencies and their political principals as a strictly hierarchical and contractual relation, in which the agent attempts to increase its discretion and autonomy by exploiting the existing information asymmetry. The principal in return attempts to hold the agent accountable for its actions through the employment of measures of monitoring, to keep the agent in check. Busuioc and Lodge’s (2017) bureaucratic reputation account of this relation is much more compound and brings forward the concept of multiple audiences to point out that ‘principal-agent’ relations are not as clear-cut as they often are conceptualized. Rather bureaucracies are constantly concerned with their image in the eyes of a large range of political

(17)

17 actors and other entities, see the way of how they hold other actors accountable and how they are held accountable themselves as functions of these considerations.

Additionally, bureaucratic reputation theory has provided the academic field of Public Administration with a wide array of novel explanations for agency and bureaucracy behaviour, that could not be explained otherwise. In this way bureaucratic reputation theory has made significant contributions to various academic debates. One example in this regard is in explaining ambiguous behaviour of agencies in their regulatory endeavours. One such contribution was made in Carpenter’s (2002) publication on FDA drug approval, in which he sets out how public saliency and organized interests have contributed as a political force attacking the FDA’s reputation, and lead to the speeding up of the approval procedures of a selection of AIDS drugs. In this sense bureaucratic reputation theory has helped to provide an insight into why some regulations are speeding up while others take much longer to approve certain drugs. Also, Carpenters (2010b) contribution on ‘institutional strangulation’ in the Obama administration’s efforts to reform financial policy is a good example. In this volume Carpenter conceptualizes bureaucratic reputational turf as an institutional veto point in the process of policy reform. Alongside the dynamics of cultural veto points and partisan deadlock, reputational concerns over threats to an organization’s regulatory/reputational turf are pointed at as one of the main reasons for policy reforms to fail (Carpenter, 2010b). Providing an insight into policy-making failure that no other perspective could properly provide.

Contributions in the field have however not only concerned themselves with cases that deal with regulatory behaviour. Also, where detection practices or standard-setting practices are concerned, bureaucratic reputation theory has contributed to our understanding. One such example has been the academic field’s assessments of how regulators’ motivations to enforce detection practices are functions of reputational concerns (Etienne, 2007). Another is Maor’s (2007) inquiry in the effectiveness of ministers’ strategies to set a scientific ‘gold standard’ for new agency policies to increase their political influence in these agencies without harming the agency’s legal autonomy. Maor (2007) found that gold-standard-setting allows ministers and other policy makers to address a large amount of reputational audiences, employing its increased political influence to address external reputational threats without endangering the policy function of these regulatory agencies.

Regarding both branches of contributions, it is only fair to say that bureaucratic reputation theory has extensively contributed in conceptual or theoretical terms as well as in

(18)

18 empirical and explanatory terms. Bureaucratic reputation theory has expanded the available theoretical toolbox to Public Administration scholars and allows us to cover a wider range of phenomena and to provide more comprehensive analyses of these cases.

2.3 Reputational Management Strategies

As was already outlined before, a central element of the bureaucratic reputation theory as emancipated by Carpenter (2010a) is that agencies, when managing their reputation towards their audiences, are embedded in a political game for legitimacy. In this game, how an agency presents itself affects (i) the degree to which the agency is able to act as an independent actor, (ii) how well it is able to carve out its own degree of agency autonomy and (iii) to which degree it is able to defend itself from external attacks (Busuioc & Rimkute, 2018). The challenges that Carpenter and Krause (2012) bring forward as consequences of this political game, and the notion that agencies adapt their behaviour and rhetoric to the pressures from their audiences, raises questions over what types of behaviour an agency can employ to project its self-image. That is, exactly how an agency can choose to present itself and manage its reputation (Carpenter & Krause, 2012). Following the accounts of Carpenter (2010a), an agency can cultivate its self-presented image along the lines of four dimensions of reputation. Each of these dimensions has its own dynamics and relevance for an organization and gives rise to how this organization aims at making itself to be understood by its audiences. The four dimensions of bureaucratic reputation are performative, technical, procedural and moral reputation (Carpenter, 2010a ; Carpenter & Krause, 2012). Organizations don not need to exclusively chose between these four to rely on only one of them – that is, the dimensions are not mutually exclusive. Rather the opposite applies: organizations are most effective in managing their reputation when relying on multi-facetted reputational strategies to reach different audiences (Busuioc & Rimkute, 2018). Each of the named dimensions of reputation will now be shortly explained.

Performative Reputation

Performative reputation, in a nutshell, looks at whether the organization at the centre of the analysis is capable of ‘getting the job done’. In this sense reputation becomes a matter of efficiency and capability and looks at whether the organization can properly execute the responsibilities and tasks it is burdened with (Carpenter & Krause, 2012). Performative reputation assesses whether the organization can effectively and competently deliver on its promises and whether it is able to achieve its ends and stated goals. In this light the ability to intimidate its audiences becomes especially important to the organization. This because the

(19)

19 organization’s interests lie with the ability to portray an image of vigour and a degree of aggressiveness when pursuing its goal (Busuioc & Rimkute, 2018). In this way organizations pursuing a strategy based on performative reputation attempt to claim legitimacy, compliance and support from its audiences for its capabilities and policies (Carpenter, 2010a). Performative reputation thus makes an agency focus on ‘deliverables that are consistent with certain policy objectives’ (Busuioc & Rimkute, 2018, p.5), which means that an agency will focus its behaviour in such a fashion that its audiences cannot help but to notice that the agency is able to regulate their specific area of competence.

Technical Reputation

Technical reputation is concerned with the degree to which the expertise of the organization in the relevant field of policy is recognized. Is the methodology that the organization employs well-informed enough? Are the people working at the organization qualified enough to be holding the regulatory responsibilities that they do (Carpenter, 2010a). In other words, does the organization have the knowledge to deal with the complex realities that it faces in exercising its tasks (Carpenter & Krause, 2012)? Technical reputation thus looks at the internal traits of the organization and whether these testify to the presence of know-how within the organization. In order to stress its technical reputation, the agency places a specific emphasis on professional signals that stress the scientific authority of the organization. This can be done through emphasizing the agency’s expertise or methodological competence (Busuioc & Rimkute, 2018). Stressing one’s technical reputation can be an exceptionally powerful tactic for an agency under threat to fend off interest groups posing a danger to the agency (Maor, 2007).

Procedural Reputation

Procedural reputation looks at the degree to which the organization is perceived by its audiences as acting in accordance with the commonly accepted and recognized rules, procedures and traditions. In other words, procedural reputation is concerned with the question whether the means by which the organization exercises its responsibilities are ‘comme-il-faut’ with regard to established practice (Carpenter, 2010a ; Carpenter & Krause, 2012). Procedural reputation looks at the justness of the procedure by which an agency comes to its behaviour, and whether this process follows the generally accepted procedures and rules. It therefore looks at the fairness of procedure, rather than at the output of the behaviour (Busuioc & Rimkute, 2018). This form of reputation thus stands aloof from questions over ‘moral’ and ‘immoral’ or ‘effective’ and ‘ineffective’, but purely looks at how the organization has come to its decisions

(20)

20 and whether these steps are in line with what is commonly accepted as the proscribed and authoritative procedure (Carpenter, 2010a ; Carpenter & Krause, 2012).

And finally;

Moral Reputation

Moral reputation, is concerned with the degree to which there exists a sense of morality that is attached to the image of the organization analysed. It assesses the degree to which the agency in case is perceived as adhering to the proper ethical standards and as being protective enough towards the consumers or users for whom it regulates (Busuioc & Rimkute, 2018). When speaking of moral reputation, what matters is the perceived compassion and respect that the agency holds towards those that are subject to its regulations (Busuioc & Rimkute, 2018). The question arises whether the organization takes proper care of the people that are dependent on the organization’s actions and whether the organization is pursuing an honourable goal, while remaining respectful to the applying ethical norms when choosing its means (Carpenter, 2010a). A number of issues that can be distinguished that moral reputation is concerned with are: does the organization follow the applying normative values in its regulations? Is it open and integer and does the organization work in the interest of the public (Busuioc & Rimkute, 2018)? In other words, is the organization honest, fair and flexible, and how does this show through in its policies? (Carpenter & Krause, 2012).

Reputational management thus is the process by which organizations adapt their

behaviour both in terms of action and rhetoric, in order to promote an image with their audiences that is in line with one or more of these forms of reputation (Carpenter, 2010a ; Carpenter & Krause, 2012). As Carpenter (2010a) already pointed out agencies balance their reliance on these dimensions to exercise their reputational management strategically. A bureaucracy’s reputation is not simply a matter of good or bad, but a matter of a good reputation in which dimension, in the eyes of what audience.

Besides the understanding of the audience, also the agency’s understanding of itself forms an important role in determining how an organization decides to present itself. This notion is expressed through the concept of reputational investment. Reputational investment as a concept captures the degree to which a bureaucracy or another organization experience that its understands its reputation to be dependent on the cultivation of one particular image towards its audiences. To the organization this image is essential to function properly in fulfilling its responsibilities and is therefore indicated as the dimension “where it matters” for an

(21)

21 organization to enhance or maintain its reputation (Busuioc & Lodge, 2016). Reputational investment has a large hand in determining the organization’s attention to its audiences. As organizations can only divide a limited amount of attention to their audiences and their external demands, they conduct into a practice of filtering the demands and critique they are faced with through their reputational considerations and investment and are accordingly prioritized along these lines (Busuioc & Lodge, 2017). Busuioc and Lodge (2016) argue reputational investment to be the link between the four forms of reputation organizations can cultivate, and what is understood in the theoretical field as reputational threat (Busuioc & Lodge, 2016, p.253).

The assessment of how agencies decide to divide their attention over the mentioned dimensions of reputation is where the study of Busuioc and Rimkute (2018) comes into scope. The authors assess the reputational management strategies of a total of 20 EU regulatory agencies over a time span of little over 20 years to investigate what factors make agencies prioritize one dimension of reputation over another. They look in this endeavour at three possible explanatory factors. First the authors look at whether over time the EU regulatory agencies lose the strong focus on technical reputation that is linked to the conception of the EU as a regulatory state. Then they look at how, as agencies mature over time, they lose their youthful regulatory fanaticism and consequently move from focussing on one dimension of reputation to increasing the spread of their attention over multiple reputational dimensions. Finally, the authors assessed the influence of ‘institutional threat’ on the differences in reputational management strategies between the different types of regulatory agencies distinguished in the article. In their findings they found clear evidence for the first two lines of thought but remained cautious in making clear-cut statements on the last one. Although they did find some evidence that advisory regulatory agencies have a tendency of stressing their technical reputation, that enforcement/supervisory agencies tend to focus more on performative aspects and that decision-making agencies tend to draw more on their technical and performative reputation (and formerly procedural too, but no longer in more recent years), they indicate that the findings in this regard are mixed as the agency-type typology is not as clear-cut as it may seem. Therefore, they suggest further research to look at the possibilities for agencies to be compared along the lines of agency category (that is f.e. regulatory agencies vs. executive/coordinative agencies) (Busuioc and Rimkute, 2018). The concept of ‘institutional threat’ is a form of the larger concept of ‘reputational threat’, which is conceptualized in the following sub-section.

(22)

22 Table 1. Summary of the dimensions of bureaucratic reputation

Form of Reputation Main Concerns Associated

Performative Reputation ‘How good is the policy output of the organization with

regards to its relevant responsibilities?’

Moral Reputation ‘Is the organization working for the public good and fair to

the people to which it is accountable?’

Technical Reputation ‘Is the organization endowed with the right expertise and

knowledge to exercise its tasks?’

Procedural Reputation ‘Does the organization follow the official and commonly

accepted procedures of policy making?’

.

2.4 The Many Faces of Reputational Threat

Reputational threat has functioned as one of the core concepts of Carpenter’s (2010a)

account of bureaucratic reputation and has continued to play an essential role in the academic field ever since. Carpenter went as far as to state that a scholar in the field bureaucratic reputation theory “… when trying to account for a regulator’s behavior, [should] look at the

audience, and look at the threats” (Carpenter, 2010a, p. 832). This call has been answered by

the academic community with a passion. An impressively large sum of authors have relied on the explanatory power of the concept to make their informed arguments (e.g. Maor & Sulitzeanu-Kenan, 2015 ; Carpenter, 2010b ; Rimkute, 2018). At the same time however, the field still lacks a clear-cut definition of what reputational threat exactly entails. That is, although the field has heavily relied on the concept, authors have mostly done so by focusing on one form of the concept at a time scope (e.g. Maor & Sulitzeanu-Keenan, 2015 ; Grøn & Salomonsen, 2017). As a result, reputational threat has been understood and used in multiple unitary ways. This sub-section collects some of the more prominently used conceptions to work towards a ‘text-book definition’ of the concept. It does so by exploring the literature occupied with reputational threat and outlining three forms in which reputational threat manifests itself. These three forms are not meant to be an exhaustive typology, and neither are they meant to signify mutually exclusive forms of the concept. After this the sub-section provides a general definition taking into account these three various forms to explain reputational threat as an overarching concept. The three manifestations of reputational threat here addressed are (i) reputational threat as institutional threat, (ii) reputational threat as saliency, and (iii) reputational threat as competition to reputation uniqueness.

(23)

23

Reputational Threat as Institutional Threat.

The first conceptualization of reputational threat here discussed is institutional threat which was mentioned shortly already. Institutional threat is derived form the idea that all organizations when exercising their function are expected to please a plethora of audiences, each of which holds its own ideas on how the agency should act. As these various audiences tend to have different ideas in this respect, the reputational demands that audiences express towards the agency tend to be conflictual and sometimes even directly contradictory. Agencies thus constantly find themselves in situations in which, to please one of their audiences, they need to disappoint another. This is where the core of the threat comes from. Such a conflict forces agencies to constantly make a consideration as to which audience to prioritize over another or to how to project ambiguity in its policies and intentions. This as the agency at all times wishes to prevent disenfranchising the audiences it needs to legitimize itself (Carpenter & Krause, 2012). In such a situation the bureaucracy thus finds itself in a reputational balancing act in which it constantly must assess and reassess the relevance of its various audiences for its reputation and the exercising of its mandate (Carpenter & Krause, 2012; Busuioc & Rimkute, 2018). This type of reputational threat is thus necessarily faced by all EU agencies, as its premise is part of the theoretical fundaments of bureaucratic reputation theory (Carpenter & Krause, 2012).

As Busuioc and Rimkute (2018) pointed out, the role that an agency fulfils within a certain substantive context, matters for what type of expectations audiences hold towards the behaviour of these agencies. In other words, a regulatory agency is faced with different audiences and accordingly different expectations then a coordinative agency. Audiences and conflicts thus vary along the lines of roles agencies fulfil. Rimkute and Busuioc pointed out how these elements may vary between the agencies with different regulatory tasks, but pointed at agency category (e.g. regulatory agencies, coordinative agencies, executive agencies etc) as a potentially more prominent dividing line between agencies institutional threats. This because as audiences and conflicts differ along the lines of agency category, so do the institutional threats that imperil agencies’ reputations (Busuioc & Rimkute, 2018). Especially in an EU context, institutional threat seems likely to be of relevance, as EU agencies are faced with a number of powerful audiences: (i) the member states and (ii) the European Commission (EC). While the agencies are heavily dependent on cooperative relations with member state authorities for the implementation of its policies and the exercising of its mandates, the agencies also significantly rely on their close relations with the EC for the necessary resources

(24)

24 and knowledge to successfully exercise its tasks (Groenleer, 2009). Institutional threat is thus dependent on agency category in the sense that a change in agency category leads to a different set of audience demands and in effect different conflicts between audiences.

How institutional threats are dealt with by bureaucracies in part depends in the reputational strengths of the relevant bureaucracy. Maor, Gilad and Bloom (2013) found that a bureaucracy’s behaviour, when faced with conflicting audiences, is largely embedded in the agency’s own understanding of its compound reputation. The authors find that ‘unified actors’, that are actors whose core mandated functions are multiple, are more likely to speak out in areas in which their reputation is already weak, as they cannot afford further reputational damage in these areas. This while they are more likely to remain silent in areas where their reputation remains firm. The authors also found that higher media coverage and more negative portrayal of the agency in the media prompts the bureaucracy more likely to respond (Maor, Gilad & Bloom, 2013). This brings us to the next form of reputational threat.

Reputational Threat as Saliency

Reputational threat as the presence of saliency is often conceptualized in academic literature as a function of media attention regarding the works of an organization. Maor and Sulitzeanu-Keenan (2015) postulated that reputational threat is a nuanced and complex concept that goes well beyond the organizations managing of its external environment. Instead, as organizations are embedded into reputational politics, attention by political principals and the wider public are of the utmost importance to these organizations. To this end the authors measured reputational threat through the strength of media coverage of organizations exercising their responsibilities (Maor & Sulitzeanu-Keenan, 2015). They found that media coverage as a ‘saliency’-reputational threat, affected an organization’s output in a corrective way. That is, when an organization would be criticized for under-regulation, the organization would increase its policy output, while for over-regulating organizations the vice versa applies. Carpenter (2002) doubles this idea of what saliency entails, when stating that a bureaucracy’s behaviour is better explained through the influence of organized interests and media coverage, rather than ideological or political motives. By conceptualizing reputational threat as negative media coverage Carpenter (2002) argues that bureaucracies are more likely to quickly approve drugs to be sold when under media scrutiny than when this form of media pressure is absent. The author would even go as far as calling the role of the media as ‘necessary’ for the approval of AIDS medication. This because media attention and organized interests ‘bullied’ the FDA until it made the drugs available for sale.

(25)

25 The degree of media attention alone however barely tells the whole story that there is to tell about ‘saliency’ as reputational threat. Not only is it important to see if the media talks about the organization, but it is essential also to look at what the media has to say. In other words: to look both at intensity and at content. The case for a focus on intensity of media coverage was made when Maor and Sulitzeanu-keenan (2013) set to investigate the interaction between visibility and reputational threat. By using a measurement of reputational threat that was pointed at the duration of media coverage to indicate that a longer duration in which an organization is covered by the media constitutes to higher intensity and vice-versa. The authors then found that media coverage intensity functions as a mediator for reputational threat. That is, its presence increases an organizations vulnerability for critique harming its reputation. At the same time Gilad, Maor and Bloom (2013) dove into the importance of the content side of saliency. Comparing the effects of media claims of ‘over-regulation’ and of ‘under-regulation’, the authors found that different kinds of content have different impacts on an organization’s reputation, leading to different considerations and forms of organizational behaviours. Claims that are of an ‘under-regulation’ nature are more likely to be acknowledged and assessed by organizations, as they pose a generally larger threat to the organization. This is argued so as they underline the organizations failure to act properly on its mandate and thus harms the organization’s image as working in the public interest (Gilad, Maor & Bloom, 2013). At the same time, claims of over-regulation offer the organization more leeway to strategize its response to the threat posed by the critique. This then allows for the organization to act differently from admitting and addressing the problem, for example by shifting the blame to other actors, or by denying the problem altogether. This while claims of over-regulation, although a reputational threat, do not portray the organization as if it has forsaken its most important responsibility: providing proper regulation to serve the public interest (Gilad, Maor & Bloom, 2013).

Saliency is thus a reputational threat that rests on the behaviour of the media, both in content and intensity terms.

Reputational Threat as Bureaucratic Competition

The final conceptualization of reputational threat here assessed is referred to as bureaucratic competition and is dependent on the causal mechanisms of the concept of

(26)

26 Reputation uniqueness refers to the ability of an organization to claim towards its audiences that the organization brings an ‘unique contribution to the public good’ (Maor, Gilad & Bloom. 2013, p. 583). The organization thus motivates its relevance to its audiences by cultivating an image as being unmissable to provide the public with a specific service. Groenleer (2014) argues that, for a bureaucracy to amass the political support and legitimacy, it needs to differentiate itself from other political entities to create its own regulatory niche. Reputation uniqueness fits in this pattern as it allows the organization to legitimize its actions within a given field by claiming its difference or uniqueness from other political entities. Carpenter (in Rimkute, 2018, p.73) exemplified the effects of reputation uniqueness of independent agencies by drawing up to lines: (i) the ability of the agency to provide expertise, scientific knowledge and efficiency in a specific field that no other actor can provide, and (ii) the agency’s acting as a ‘guardian for social welfare’ by providing policy output that is embedded in wider moral implications. Rimkute (2018) goes on to show how ‘risk regulators’, belonging to the first line Carpenter set out, tend to cultivate more their reputation to fit their image as experts and scientific authorities, while ‘social value guardians’ tend to cultivate their reputation more to fit their protective roles towards social values.

Reputation uniqueness has various important functions for organizations. One example is how reputation uniqueness lies at the fundament of organizations making regulatory

jurisdictional claims. Maor (2010) holds that a threat to an organizations reputation uniqueness,

can motivate the organization to use their reputation uniqueness to expand their informal mandate by claiming its regulatory authority in different policy fields (Maor, 2010). The claiming of additional regulatory turf by organizations is a process that starts with the organization claiming its additional territory and ends with the formal recognition of its control. Maor (2010) claimed that two threats to the reputation uniqueness of an organization could pressure an organization to pursue a regulatory jurisdiction claim over new technology entering the policy field. These are (i) the new technology turns out to have the potential to bring serious harm to society if it remains unregulated, and (ii) when a rival regulatory organization fails to regulate the new technology even when it officially belongs to its mandate. The study showed that (although for novel areas to be regulated certain thresholds do exist), the jurisdictional claims of a competitive regulatory actor moving into another actor’s official mandate can create pressures to the latter to start regulating this new regulatory area too as its reputation uniqueness is under threat. That an organization can be threatened in its reputation uniqueness, and that this can lead to problems for an organization to fulfil its task is confirmed by Busuioc (2016),

(27)

27 who shows how the overlapping of regulatory turf between national policing authorities and Europol has imperilled Europol’s ability to properly cooperate with other organizations. From a reputation uniqueness perspective inter-agency cooperation is tricky business as it allows different organizations to move into another organization’s regulatory jurisdiction. If cooperation can happen without organizations forming a threat to one another’s reputation uniqueness we speak of reputation-enhancing cooperation as it allows organizations to better exercising their tasks, without imperilling their image as a unique organization in the eyes of their audiences. When however, the organizations feel that the envisioned ‘cooperation’ in fact is an attack on their regulatory turf, they will see cooperation as a threat to their reputation uniqueness and we speak of reputation-depleting cooperation. As Busuioc’s (2016) and Maor’s (2010) accounts thus show organizations are embedded in a political strife for regulatory and reputational turf with other agencies to secure their reputation uniqueness and with that the image carrying their legitimacy. This means that if another actor copies the organization’s function in a certain policy (so both its performative and its substantial side), that the organization is experiencing a threat to its reputational uniqueness. This reputational threat is referred to in this thesis as bureaucratic competition. This form of reputational threat differs from institutional threat as bureaucratic competition only occurs when another agency attempts to expand its turf at the cost of another agency, effectively questioning the seconds reputation uniqueness, while institutional threat is dependent on audiences disagreeing with one another and thus is embedded in the organization’s agency category.

(28)

28 Table 2. Summary of the forms of reputational threat.

Form of Reputational Threat

Concise Explanation Dimensions to the Concept

Reputational Threat as Institutional Threat

The cultivation of a reputational image by an organization is imperilled through the

disagreement in demands between essential actors within the

organization’s audiences.

- Different roles for the agency means

different audiences

Reputational Threat as Saliency

The cultivation of a reputational image by an organization is imperilled through content or intensity-wise media attacks on the reputation of the relevant organization. - Content - Intensity Reputational Threat as Bureaucratic competition

The cultivation of a specific reputational image of reputation uniqueness of an organization is endangered by the moving in of another organization into the organizations regulatory jurisdiction/turf.

Seen the findings of the investigation into these various forms of reputational threat, the following commonalities can be found on what these different forms understand reputational threat to be. In each of the forms of reputational threat, the deliberate cultivation of a certain reputational image of an organization is disrupted and problematized by an external development. This external development can come in multiple forms, whether it comes from another organization, the media or actors amongst the organization’s audiences. The pressure exerted by these external developments then force the organization to alter either the magnitude or the content of its management strategies to reinforce or alter the image it attempts to cultivate with its audiences. This regarding, the thesis adheres to reputational threat as:

‘An external development that disrupts an organization’s deliberate cultivation of a specific reputation, and that an organization needs to adapt to through further strategic reputational management by the organization in order not to experience negative consequences from that specific development.’

(29)

29 2.5 Theoretical Framework

A Reputational System

This sub-section uses the theoretical concepts and mechanisms as found in the foregoing literature review to provide an oversight of how the dependent and independent variables to the thesis are interrelated. Following up on the advice of Carpenter (2010a, p.832) to “…look at the audience and look at the threats.” the thesis wishes to provide a full oversight of how the plethora of introduced concepts in this section interact. It does so by looking at the proliferation of reputational management strategies as a process of input, throughput and output. In order to make this effort a bit easier to oversee, the sub-section uses the procedural insights of Easton’s (1965) theory of political systems to structure this endeavour. Easton’s (1965) theory is considerably different from the bureaucratic reputation theory. It hold on to different units of analysis (states instead of organizations), mostly focusses on material factors rather than abstract concepts such as ‘reputation’ and was in general established to explain different phenomena. The mechanisms that underlie the two theories are in core however quite similar, as will be now explained.

In Easton’s (1965) systems theory, political systems interact with their environment through a feedback-loop structured process wherein demands and support of the ‘environment’ are assessed by the political system and are used by the political system to determine future policy and behaviour of the political system. A state takes in political demands and support from its environment (most of the time a government’s constituency), which it then processes, prioritizes and uses to create new policy. In this sense demands and support from the systems input, the systems consideration forms the throughput and the renewed behaviour and policies form the organizations output. The new output is then reassessed by the political environment and changes the degree to which the environment supports the political system (that is, recognizes it as legitimate) and what the environment demands from the political system. Easton (1965, p. 32) conceptualized this relationship as followed:

(30)

30

Figure 1. Easton’s (1965, p.32) political systems theory.

Easton (1965) includes a concept of stress, that plays into this model as an additional factor, disrupting the stability of the system and puts the political system into crisis (Easton, 1965). The mechanisms by which the concepts of bureaucratic reputation theory interact are relatively reminiscent with the way Easton’s (1965) concepts interact in his model. The environment of the organizations is formed by its relevant audiences. The inputs the audiences bring forward come indeed in the form of political support that legitimizes the organization and of demands, being the reputational images and expectations, the audiences hold towards the organization. Like in the political systems theory, the reputational demands and support of the relevant audiences are assessed and prioritized by the organization and then transformed into fitting output responses. The conversion of input into strategic output in this system is ruled by reputational investment. As Busuioc and Lodge (2016 ; 2017) pointed out, reputational demands are interpreted and prioritized along the lines of what reputation ‘matters’ to the organization and follow the line of what type of reputation organizations are invested in. In this sense reputational investment is a filter to audience demands and external developments. Reputational investment is thus a process through which an agency assesses the inputs it receives from its audiences and should be seen as part of the throughput phase (Busuioc & Lodge, 2016 ; Busuioc & Lodge, 2017).. The decisions and actions that the organization then forms are specific outputs that match the input, in the bureaucratic reputation theory in the form of reputational management strategies. The bureaucracy’s reliance on specific forms of reputation, be it moral, procedural, technical or performative are used to structure behaviour and rhetoric the organization produces to make sure these fit the reputational demands of their audiences (Carpenter, 2010a). In this sense the behaviour and rhetoric of the audiences are best understood as the system’s output.

Talking Under Threat. A mixed-methods approach to the effects of reputational threat on EU agencies' reputational management strategies