The effect of whistleblowing as a reputational threat on the United Nations’ communication strategy

(1)

The effect of whistleblowing as a reputational threat on the

United Nations’ communication strategy

Name: Manon Dees

Student number: 1382047

15 July 2019

Leiden University

Master Public Administration

Track: International and European Governance

Faculty of Governance and Global Affairs

Thesis supervisor: Dr. J. Christensen

Second reader: Dr. D. Rimkute

(2)

i

Acknowledgements

This thesis marks the end of the master’s program Public Administration, with a specialization in International and European Governance, at Leiden University. I would like to use this preface to thank some people, who have helped me through the process of writing this thesis. First and foremost, I would like to thank my thesis supervisor, Dr. Johan Christensen. He guided me through the process from the beginning to the end, provided me with fair and thorough suggestions, and was rightfully critical where needed. I would also like to thank the second reader, Dr. Dovile Rimkute, for taking the time to read and assess this thesis.

Following a second master’s program, including writing a second master’s thesis, many people called me crazy. However, I am glad that I had this opportunity and gained many new insights in this past year, both in terms of academic knowledge as well as where my interests lie and what I am passionate about. The people who called me crazy are also the ones that always have confidence in me and allow me to follow my intuitions. Thank you for that.

(3)

ii

Abstract

An increasingly popular way to report misconduct or wrongdoing within organizations is through whistleblowing mechanisms, which receive growing attention from both scholars and practitioners. On the one hand, whistleblowers provide valuable inside information that organizations may use to improve their policies and services. On the other hand, whistleblowing may pose a significant threat to an organization’s reputation as it often contains harmful information about the organization’s illegal, immoral or illegitimate practices.

The tension between whistleblowing as a contribution to an organization’s effectiveness and whistleblowing as a reputational threat may lead to seemingly irrational choices concerning whistleblower disclosures in terms of disproportionate and defensive response strategies. The organizational response varies from whistleblower case to whistleblower case. The conditions that shape an organization’s response strategy, however, are relatively unknown. In order to gain more insight into this latter aspect, the main research question of this thesis is: “Which

conditions explain how the United Nations responds to whistleblowing as a reputational threat in their communication activities?”.

Based on empirical analysis of newspaper articles and other relevant policy documents, this thesis argues that reputation-seeking behavior may explain the United Nations’ response to whistleblowing. The theoretical framework provided in the bureaucratic reputation literature is useful to assess whistleblowing as a reputational threat. However, whistleblowing poses a rather specific reputational threat, as the process evolves following a two-stage process. Initially, whistleblower allegations pose a threat to the UN’s moral reputation, as they disclose information about the organization’s immoral or unethical behavior. Then, if the UN responds inappropriately by violating whistleblower protection rules and procedures, the allegations grow out to be a threat to the organization’s procedural reputation.

However, a single set of conditions to explain the variation in these response strategies seems to be insufficient. Rather, the relative threat that the allegations pose to the organization’s reputation, heavily shaped by the context in which the disclosures take place, seems to determine the UN’s response strategy. Important determinants seem to be whether the whistleblower disclosures have been preceded by similar allegations, or whether there have been other events in parallel that may reinforce the pressure on the UN to respond.

(4)

iii

List of tables

Table 1: Agency response strategies ... 16

Table 2: Covariational table ... 25

Table 3: Volume of document analysis ... 29

Table 4: Operationalization of response strategies ... 29

Table 5: Operationalization of conditions ... 30

(7)

vi

List of abbreviations

AU African Union

CAR Central African Republic

DPKO Department of Peacekeeping Operations GAP Government Accountability Project HRC Human Rights Council

MINUSCA Mission multidimensionnelle intégrée des Nations unies pour la stabilisation en

Centrafrique

MINUSTAH Mission des Nations Unies pour la stabilisation en Haïti OECD Organization

OHCHR Office of the High Commissioner for Human Rights OIOS Office of Internal Oversight

UN United Nations

UN JIU United Nations Joint Inspection Unit

UNAMID African Union - United Nations Hybrid Operation in Darfur UNDT United Nations Dispute Tribunal

UNHCR United Nations High Commissioner for Refugees

UNMIK United Nations Interim Administration Mission in Kosovo UNSC United Nations Security Council

(8)

1

1. Introduction

1.1 Problem definition

Employees or former employees are often the first to report malpractice or illegal activities in their organizations. An increasingly popular way for them to do so, is through whistleblowing mechanisms (Thomson Reuters, 2014). Whistleblowing is “the disclosure by organization members (former or current) of illegal, immoral, or illegitimate practices under the control of their employers, to persons or organizations that may be able to effect action” (Miceli & Near, 1985, p. 4). As such, it is considered an increasingly important means to combat corruption and other illegalities in both public and private organizations (Martel, 2017).

Whistleblowing fits within the landscape of transparency and anti-corruption efforts that is increasingly advocated for by both citizens and government organizations (Brown, Vandekerckhove & Dreyfus, 2014). Whistleblowers are important sources of information to report bureaucratic or corporate wrongdoing as they possess specific knowledge or expertise (Culiberg & Mihelic 2017). They can be considered ‘third-party informants’, on which organizations rely to detect non-compliant behavior (Etienne, 2014). More specifically, they are “the source of valuable information that neither the government nor the public can get from oversight systems” (Apaza & Chang, 2011, p. 113).

The role of whistleblowers to disclose misconduct in organizational settings is gaining prominence among both practitioners and scholars. Employers and governments actively encourage whistleblowers to speak up and report systematic wrongdoing in order to protect the public interest (Council of Europe, 2015). Multinational companies, such as Heineken, Shell, and Volkswagen have introduced whistleblower guides to encourage the reporting of misconduct within their organizations. Also, multi-stakeholder initiatives such as CleanGovBiz, facilitated by the OECD, actively support governments to encourage the reporting of corruption, fraud or misconduct (OECD, 2012). The importance and growing prominence of whistleblowing also appears from a growing scholarly interest in whistleblowing mechanisms (e.g. Dyck, Morse & Zingales, 2010; De Graaf, 2010; Lewis & Trygstad, 2009), and a growing body of legislation on whistleblower protection (e.g. OECD, 2016; European Commission, 2018).

(9)

2 Despite their growing prominence, organizations still seem somewhat uncomfortable in dealing with whistleblowing or disclosure of information that may harm their reputation (Martel, 2017). A case in point is the United Nations (UN). With its UN Convention against Corruption, the international organization takes a firm stance on the importance of whistleblower protection. It encourages signatories to the Convention to adopt legal provisions to protect whistleblowers in different contexts. Protection to the whistleblower may include, for example, psychological support, the transfer or relocation of the whistleblower within the organization or to another organization, and the institutional recognition of reporting (Transparency International, 2013). However, the UN does not have a proven track record regarding the protection of its own whistleblowers. Multiple examples exist of cases in which (former) UN officials faced retaliatory measures by the organization after they reported wrongdoing in their own working environment (Government Accountability Project [GAP], 2014; Martel, 2017).

As argued in this thesis, the discrepancy between the position of the UN on whistleblowing in the international environment and its position regarding its own whistleblowers can be explained by bureaucratic reputation theory. This theory argues that bureaucracies are politically conscious organizations that actively seek to protect their reputation in a specific polity (Carpenter, 2001; Maor & Sulitzeanu-Kenan, 2013; Rimkute, 2018). Consequently, organizations may adapt their behavior strategically to establish or maintain a good bureaucratic reputation if they experience a reputational threat (Maor, 2015). Detection of non-compliant behavior through whistleblowing mechanisms may form a significant reputational threat for organizations (Etienne, 2014). In responding to these threats, organizations may choose from different strategies in order to protect their distinct reputation or to mitigate negative consequences. However, bureaucratic reputation literature lacks clarity about the conditions under which organizations are likely to choose a particular strategy to respond to reputational threats.

1.2 Research question and objective

How organizations respond to reputational threats varies heavily across different organizations and different contexts. Organizations may choose from different strategies, including, but not limited to, regulatory talk or strategic silence (Maor, Gilad & Bloom, 2013), blame avoidance or blame shifting (Hood, 2011), deploying scientific risk assessments (Rimkute, 2018), or accepting responsibility (Gilad, Maor & Bloom, 2013). Gilad, Maor & Bloom (2013) arrive at

(10)

3 a set of conditions under which regulatory agencies tend to choose for either one of these strategies. However, they argue that it is unclear to what extent their conclusions can be generalized to other types of agencies and contexts (Gilad, Maor & Bloom, 2013, p. 474).

Therefore, the objective of this research is to provide an explanatory study on the conditions that may influence an international bureaucracy’s choice for a strategy to respond to whistleblowing as a reputational threat. This thesis will focus on the organizational context of the United Nations to answer this question. To this end, the research question central to this thesis will be: “Which conditions explain how the United Nations responds to whistleblowing

as a reputational threat in their communication activities?”.

In order to arrive at an answer to this research question, the following sub-questions will be addressed:

• To what extent is whistleblowing considered a reputational threat by the United Nations?

• What strategies does the UN pursue in response to whistleblowing as a reputational threat?

• Which conditions explain the UN’s response to whistleblowing?

1.3 Research design and methods

In order to answer the research question, this thesis uses a small-N comparative research design complemented with causal process-tracing. Four whistleblower cases from the context of the UN have been selected that each represent a different response strategy, which have been identified in the literature review. The cases that are selected are that of Aicha Elbasri, Emma Reilly, James Wasserstrom, and Anders Kompass. Respectively, they represent the response strategies of silence, denial of allegations, blame shifting, and admitting responsibility. Due to the comparative nature of this study, it is possible to identify a covariation between the conditions and the response strategies by controlling for other possible explanations. The within-case analysis enables us to further substantiate these claims and allows us to take into account context-specific explanations.

The empirical analysis relies on a qualitative document analysis of both primary and secondary documents: policy documents, UN press releases, investigative reports, and newspaper articles.

(11)

4 A total of 60 documents have been analyzed, although the number of documents available per case varied substantially. The documents have been coded for the conditions that have been identified in the theoretical framework. Moreover, the qualitative nature of this study allows us to take into account the cross-case variation in the content of the allegations in terms of their severity and salience.

1.4 Scientific relevance

Whistleblowing as a way to report organizational misconduct receives growing scholarly attention. However, research has thus far focused predominantly on the personal characteristics of whistleblowers (Bjørkelo, Einarsen, Nielsen & Matthiesen, 2011; Vandekerckhove, 2006), the intentions of individuals to blow the whistle (Lavena, 2016; Mesmer-Magnus & Viswesvaran, 2005), and the legal and institutional frameworks of whistleblower protection (Loyens & Vandekerckhove, 2018). However, the question of how organizations respond to whistleblowing has been relatively understudied. The literature on how organizations respond to whistleblowing focuses predominantly on responses in terms of policy change (Apaza & Chang, 2011; Johnson & Kraft, 1990), or in terms of the retaliatory measures taken against the whistleblower (Miethe & Rothschild, 1999). Whistleblowing literature has, for example, shown how organizations dismiss the allegations and suspend the whistleblowers involved (Brandt & Svendsen, 2013). However, a more coherent framework to understand organizations’ responses to whistleblowing is missing. This thesis will add to this literature by studying how the United Nations in particular responds to various cases of whistleblowing.

Moreover, this thesis adds to bureaucratic reputation literature in two ways. First, it contributes to bureaucratic reputation literature by focusing on whistleblowing as a specific reputational threat. Bureaucratic reputation theory has focused on how organizations respond to negative publicity and public allegations in general (Gilad, Maor & Bloom, 2013; Maor, Gilad & Bloom, 2013; Maor & Sulitzeanu-Kenan, 2013) and how organizations may deal with unsolicited third-party information (Etienne, 2014). However, whistleblowing as a specific reputational threat has remained underexposed, despite the fact that whistleblowing has often been associated with reputational damage (Near & Miceli, 2016; Tsahuridu, 2011). Although whistleblowing cases can be considered a specific category of public allegations, the fact that they stem from (former) employees and their relative complexity makes them interesting to theorize and study in isolation.

(12)

5 Second, this thesis adds to bureaucratic reputation literature by studying the conditions that shape an organization’s response to reputational threats. Research on how bureaucratic organizations respond to reputational threats is relatively extensive. In some cases, organizations choose either to remain silent or to actively counter negative publicity (Maor, Gilad & Bloom, 2013). In other cases, organizations may avoid or shift the blame when confronted with reputational threats (Bovens, ‘t Hart, Dekker & Verheuvel, 1999; Hood, 2011). However, the conditions under which organizations choose a particular strategy remain unclear and deserve to be studied further. This thesis contributes to this gap by assessing the conditions under which organizations are likely to respond to reputational threats, and whistleblowing more specifically.

1.5 Societal relevance

In terms of societal relevance, this research contributes to the debate on transparency and anti-corruption efforts that is increasingly important to public organizations. Whistleblowing is one of the cornerstones of organizational transparency and accountability. At the same time, as this thesis argues, it is also often at odds with reputation-seeking behavior. As organizations seek to protect their reputation, they may take a rather aggressive stance towards the whistleblower, which may come at the cost of transparent and accountable governance. By shining light on this tension between reputation-seeking behavior and transparency, this thesis raises an important societal issue.

Moreover, this thesis illustrates how the organizational response is crucial for the impact of the whistleblower allegations and for the actual measures that are being taken to address the issue. By providing insight into the conditions under which whistleblowing is likely to result in a certain organizational response, actors may decide to facilitate certain conditions or not. For example, if it turns out that a threat to the organization’s resources is an important condition for organizations to admit their misconduct and the implement corrective action, the whistleblower or other actors involved may want to search for ways in which they can exert pressure on the organization’s resources. They may, for example, threaten to suspend their financial contribution to the UN, or demand that troops are withdrawn from peacekeeping missions.

Finally, whistleblowing within the context of the United Nations has recently received heightened attention from practitioners. In 2016, the organization faced severe public criticism

(13)

6 after the publication of a report, which showed that fear of retaliation ranked high among UN staff (Bartsiotas & Achamkulangare, 2016). This, in turn, has resulted in staff feeling uncomfortable reporting suspicions of misconduct or fraud, which may lead to more misconduct being left unreported. Therefore, gaining insight into the conditions that shape an organization’s response may provide more insight as to why the organization retaliates against its whistleblowers, which could be a valuable insight to ultimately break this vicious cycle.

1.6 Structure

In order to answer the main question of this thesis, a thorough understanding of bureaucratic reputation literature is required. Therefore, a review of this literature will follow in chapter 2, in which the core theoretical concepts and ideas will be discussed. Based on this literature review, several expectations are outlined that are used as guidance for the analysis. In chapter 3, the research design and methodology will be discussed, together with their limitations. Subsequently, chapter 4 will provide a brief overview of the cases. In chapters 5 and 6, the main results of the empirical study will be presented. Chapter 5 will focus on how whistleblowing appears as a reputational threat within the context of the United Nations. Chapter 6 will elaborate on the response strategies that the UN pursues in responding to whistleblowers, and the conditions that have shaped the UN’s response to whistleblowing cases will be discussed. Finally, in chapter 7, the main conclusions of this thesis will be presented, as well as a discussion and recommendations for further research.

(14)

7

2. Theoretical framework

2.1 Bureaucratic reputation: an introduction

Guided by bureaucratic reputation theory, this thesis will study which strategies for responding to reputational threats organizations may deploy and under which conditions they may do so. In order to answer the main question, a thorough understanding of the academic literature is required. Therefore, this section will first introduce the key concepts and theoretical ideas that are the foundations of bureaucratic reputation theory.

Rather than operating in a political vacuum, administrative organizations operate in an environment that is characterized by a multiplicity of stakeholders, such as elected officials, the media, interest groups, and citizens. When steering through this political environment, bureaucracies typically face administrative challenges related to the support for their activities, their “enemies and potentially disaffected supporters”, and their consistency and flexibility (Carpenter & Krause, 2012, p. 26). In order to adequately respond to these challenges, public administrations heavily rely on their distinct reputation (Ibid.). As such, a bureaucracy’s distinct reputation is an important political asset that is used to accumulate power and autonomy vis-à-vis its political overseers and other stakeholders (Carpenter, 2001).

The central tendency of bureaucratic reputation theory stands in contrast with principal-agent approaches to understand the autonomy and power of bureaucracies (Carpenter & Krause, 2015). Whereas in principal-agent theories it is assumed that the power of bureaucracies depends on the information advantage that a bureaucratic agency has, reputational accounts of organizational behavior assume that reputation is an important source of power (Maor, 2015). As such, reputational incentives increasingly shape bureaucratic politics (Carpenter & Krause, 2015). Moreover, complementary to rational accounts of organizational behavior, bureaucratic reputation theory provides an alternative model to understand regulators’ behavior. Its theoretical framework allows for greater complexity of human motivations and provides a more ‘psychologically realistic’ model of public organizations, taking into account the complexities of politics and policy making (Carpenter, 2010, p. 43).

The theoretical framework on organizational reputation started to develop after Carpenter’s (2001) seminal work on how regulators are rational and politically conscious actors that seek to carefully forge and protect their unique reputations, which is the main premise of

(15)

8 bureaucratic reputation theory. In this context, organizational reputation may be defined as “a set of beliefs about an organization’s capacities, intentions, history, and mission that are embedded in a network of multiple audiences” (Carpenter 2010: 33). Audiences in this context may include political and judicial authorities, interest groups, professional or scientific experts, the media, clientele groups, and citizens (Carpenter, 2002; Carpenter & Krause, 2012).

Reputation is not fixed or formalized, however. Instead, Carpenter (2010, chap. 1) emphasizes that reputation is merely a symbolic construct that is subject to various interpretations. Reputation, therefore, remains a rather abstract concept, but in its essence, it is a form of organizational identity or image. Ideas about the organization’s identity and image can be both internal, how the organization perceives itself, and external, how it is perceived by outsiders. As such, an organization’s reputation can affect how its members behave and how it interacts with external actors. An organizational reputation, then, is an important aspect of the organization’s identity, and how it projects this externally (Carpenter, 2010, chap. 1).

Carpenter (2010, chap. 1) distinguishes between four dimensions of organizational reputation: performative, moral, technical, and procedural dimensions. Performative reputation relates to the quality of the organization’s decision-making and its capacity to effectively achieve its objectives. The most pertinent question here is whether ‘the agency can do the job’ (Carpenter & Krause, 2012, p. 27). Moral reputation relates to the general recognition of the organization as being moral and whether it has a culture of ethical behavior and transparency. Technical reputation refers to the organization’s expertise and capacities that are necessary to deal with a specific subject. Procedural reputation relates to the capacity of an organization to follow generally recognized legal procedures of decision-making (Carpenter, 2010, chap. 1).

An important feature of these dimensions is that they represent a tradeoff. An organization does not have the capacity and resources to pay equal attention to all four of these dimensions. Emphasizing one dimension may come at the expense of another dimension; by paying attention to one dimension, it automatically neglects another (Carpenter, 2010; Carpenter & Krause, 2012). For example, an organization may prioritize to adhere to procedural rules of decision-making so that it neglects efficiency, which is at the core of its performative reputation. Considering that different audiences may prioritize different dimensions, focusing on one dimension may please a certain audience, while leaving another audience dissatisfied. And as the external support for an organization’s existence and activities is derived from its

(16)

9 audiences’ behavior, the organization needs to strike a balance between the various dimensions and audiences.

The organizational challenge, then, is for organizations to meet the expectations and beliefs of multiple audiences simultaneously. External stakeholders do not constitute a coherent and united audience, but are diverse in their opinions and beliefs about the organization’s performance. Satisfying the expectations of one audience could result in the dismay of another audience (Carpenter & Krause, 2012). The decision for an organization to consolidate its reputation with one audience over the other may be based on the assessment of the relative threat posed by that particular audience (Gilad & Yogev, 2012).

An organization may significantly benefit from establishing and maintaining a good reputation. Bureaucratic reputation theory draws on the idea that bureaucracies continuously pursue external support, which may enhance their autonomy vis-à-vis their political overseers and other stakeholders (Carpenter, 2001, 2010; Carpenter & Krause, 2012; Maor, Gilad & Bloom, 2013). A well-established reputation may enable the organization to persuade elected officials or political overseers to defer to the preferences of the organization (Carpenter, 2001) and as such may benefit the administrative discretion within the organization (Carpenter & Krause, 2012). Moreover, organizational reputation can also be used to protect the organization from political attacks or hostile external audiences (Carpenter, 2002; Hood, 2011). The third section of this chapter will discuss this latter aspect more in detail.

2.2 How do administrations manage their reputation?

Organizations are engaged in a continuous process of forging and maintaining their reputation. The aim of this section is to give a general overview of how organizations may establish and subsequently protect their distinct reputations.

Gilad & Yogev (2012) define three reputation-management strategies that organizations may deploy. First, an organization’s reputation depends on its functional uniqueness (Carpenter, 2001; Maor, 2010). Therefore, organizations attempt to establish their reputation by seeking a niche role that is distinct from other organizations in that specific policy field (Gilad & Yogev, 2012, p. 322). Second, organizations actively seek to shape public expectations of their roles and activities. As the gap between their self-perception and the public expectations narrows, it

(17)

10 is more likely that the organizational reputation will be maintained (Gilad & Yogev, 2012, p. 323). Third, organizations can protect their reputation by striking a balance between the different, sometimes opposing, dimensions of their reputation. As was mentioned in the previous section, different audiences may expect and prioritize different dimensions of an organization’s reputation. As such, organizations need to identify threats posed by their different external audiences, and need to find a balance between them (Gilad & Yogev, 2012, p. 324; Carpenter & Krause, 2012).

Moreover, organizations may manage their reputation by strengthening account-giving and account-holding efforts (Busuioc & Lodge, 2017). An organization that is accountable is more likely to generate support, as it can systematically keep track of and manage its audiences’ expectations and perceptions. As organizations are more likely to invest in areas that relate to their core competencies and resources for reputation (Carpenter, 2010), it is in those areas that organizations are more likely to demonstrate accountability behavior. It is in those areas related to the organization’s core competencies where it matters most whether an organization’s reputation is being maintained or harmed (Busuioc & Lodge, 2017).

Another strategy for organizations to manage their reputation may be to avoid or shift the blame. Blame avoidance has been theorized by Hood (2002, 2007, 2011) and has been developed independently of bureaucratic reputation theory. Blame avoidance refers to strategic behavior that politicians or regulators may deploy to avoid the blame rather than claim political credit (Hood, 2007). The fundamental difference between bureaucratic reputation theory and blame avoidance theory is that the former assumes that organizations actively pursue a positive reputation, whereas the latter assumes that organizations focus on avoiding and managing blame (Gilad & Yogev, 2012).

2.3 Reputational threats

Bureaucratic reputation is fragile, however. Bureaucratic reputation theory assumes that an organization’s reputation will be maintained as long as it can provide the expected quality of services and effectiveness of outputs over time, and as long as it satisfies its diverse audiences in doing so (Maor, 2015, p. 24). However, in the pursuit of preserving and strengthening their reputation, bureaucracies may face claims or behavior that undermine their identity or image. Consequently, these claims or behavior may harm an organization’s distinct reputation and can

(18)

11 therefore be considered reputational threats, which can cause a possible loss of an agency’s reputational capital (Maor, 2015, p. 20).

An important feature of reputational threats is that they are “man-made products of social interaction and communication” (Power, Scheytt, Soin & Sahlin, 2009, p. 302). Reputational threats may emerge from criticism regarding an agency’s performance on issues related to their core functions. This may be the result of various external or internal forces that put pressure on an agency’s reputation. This could range from a change in the agency’s leadership to globalization pressures, and from organizational scandals or accidents to engaging in moral or scientific debates (Maor, 2015, p. 24). Carpenter (2010b) points out that agencies are often predominantly concerned with protecting their ‘turf’, that is an agency’s distinctive jurisdiction or mission (Busuioc, 2016; Maor, 2010). Criticism that organizations receive regarding their poor performance may threaten their turf and resources. As such, these man-made reputational threats may significantly harm an organization’s reputation and consequently, its survival (Maor, 2015).

Criticism regarding an agency’s performance may stem from its various audiences. Due to the multiplicity of audiences, different threats to an agency’s reputation may emerge from these different audiences (Maor, Gilad & Bloom, 2013). As mentioned in the first section of this chapter, a major challenge for public organizations is to meet the expectations of various audiences at the same time. By satisfying one audience, another audience automatically is neglected, which may carry a reputational risk. The question which audience to prioritize is therefore a question of which reputational risk to take. As summarized by Maor (2015): “The prioritization of reputational risks may involve a consideration of the relative risk to society, the relative risk to the most relevant external audience, the (political) cost of damaged reputations, and the extent to which the agency is willing to expose itself, by way of policy intervention, to reputational loss” (p. 21).

As a logical response, organizations will seek to mitigate reputational threats, considering the ‘asset status’ of reputation (Carpenter, 2010, p. 60). Considering the fact that organizational reputation can be used as a political asset, by generating public support and autonomy and by protecting the agency from political attacks, organizations will actively seek to shield themselves from threats to their distinct reputation. However, factors that may threaten an

(19)

12 organization’s distinct reputation are often unstable and unpredictable, which makes it difficult for an organization to manage reputational threats.

2.4 Organizational responses to reputational threats

Since organizations consider their reputation a valuable political asset, they are likely to carefully protect their distinct reputation by responding to claims or events that may damage it. In order to maximize support and mitigate negative feedback, organizations carefully decide which strategies to deploy as a response to such reputational threats (Gilad, Maor & Bloom, 2013). As summarized by Maor (2015), when facing a threat to their distinct reputation, “reputation-sensitive agencies are adaptive, strategic, and sometimes even opportunistic actors” (p. 24).

2.4.1 Response strategies

The literature on how organizations respond to reputational threats is relatively elaborate and focuses on various mediums through which organizations may respond. Organizational responses to reputational threats have been assessed in terms of decision-making processes (Carpenter, 2002; Gilad & Yogev, 2012; Maor & Sulitzeanu-Kenan, 2013), jurisdictional claims (Maor, 2010), inter-agency cooperation (Busuioc, 2016), agency output (Maor & Sulitzeanu-Kenan, 2015), and their scientific conduct (Rimkute, 2018). However, this thesis will focus on how organizations may respond to reputational threats in terms of their communication activities. Therefore, this section will provide an overview of the response strategies that organizations may choose from in terms of their communication.

Agencies’ response to reputational threats has been explored in terms of their communication activities by Gilad, Maor & Bloom (2013) and Maor, Gilad & Bloom (2013). The role of communication in relation to reputation management has been relatively understudied (see for an exception Busuoic & Rimkute, 2019). Maor, Gilad & Bloom (2013) aim to fill this lacuna by considering communication as a central aspect of public agencies’ reputation management and by investigating how communication can play a role in efforts to mitigate reputational threats. Similarly, Gilad, Maor & Bloom (2013) study how the content of public allegations affects organizations’ communications strategies. Both studies emphasize that it is the relative threat to its reputation that shapes an agency’s communication strategies.

(20)

13 Based on a case study into the Israeli banking regulator’s response to public allegations, Maor, Gilad & Bloom (2013) distinguish between regulatory talk and strategic silence as different possible response strategies that organizations may choose from when facing a reputational threat. Organizations may opt to remain silent when they are faced with external judgements of their performance for strategic reasons. Strategic silence could bring benefits to the organization, when the release of information about problems could potentially harm the organization or even the bigger system. For example, in the case of the financial sector, communication about failures within central banks may eventually lead to collapse of the financial system. Instead, when organizations choose to remain silent, they send a signal to external audiences that there is no reason to worry and that it is confident in its own position. Moreover, silence may send a signal that the organization is currently working on the problem or may be used to stifle crisis situations (Maor, Gilad & Bloom, 2013, p. 585).

Just as choosing to remain silent, an organization may choose to actively respond when faced with external judgements of its performance. In order to protect and maintain their reputation, organizations will actively seek to manage the expectations of external audiences. Maor, Gilad & Bloom (2013) expect that an organization’s communication activities will play an important role in this. Communication efforts may shape the way in which external audiences judge their performance and their success or failure. Therefore, rhetoric may be used by organizations to “increase the regulator’s popularity and protect it from audience-based pressures” (Maor, Gilad & Bloom, 2013, p. 585).

Although Maor, Gilad & Bloom (2013) consider these response strategies as a ‘binary choice’ between regulatory talk and strategic silence, Gilad, Maor & Bloom (2013) further specify the different kinds of response that are possible next to the strategy of nonresponse. They distinguish between denial of allegations, blame shifting strategies, and accepting responsibility. Denial of allegations entails the denial of harmful events or failures, their significance, or the justification of harmful events in light of public values and interests. Blame shifting and accepting responsibility, on the other hand, are both part of problem admission strategy. They both involve conceding that a harmful event has happened rather than denying it, but differ in their intention. Whereas blame shifting entails shifting away the responsibility to others, admitting responsibility means that the organization publicly acknowledges that the mistakes can be attributed to them (Gilad, Maor & Bloom, 2013).

(21)

14 In short, organizations selectively respond to various reputational threats emerging from different external audiences (Maor & Sulitzeanu-Kenan, 2015; Rimkute, 2018). As organizations are judged by different external audiences, “they have to play different reputation-promotion games” (Rimkute, forthcoming, p. 6). This results in uneven attention paid to different reputational threats stemming from different external audiences. This may result in different strategies being deployed in different situations. Organizations are likely to respond first to claims or events that are most threatening to their distinct reputation, that is, they respond to an assessment of the relative threat to their reputation. This process is referred to as ‘selective response’ (Maor & Sulitzeanu-Kenan, 2015, p. 34) and underlines the idea that organizations are rational and politically conscious actors able to assess the seriousness of reputational threats. Depending on the assessment of the relative threat to their distinct reputation, organizations will decide on how to respond to external judgements of their performance (Gilad, 2012).

2.4.2 Conditions for response

As appears from the previous section, four different response strategies can be identified based on the literature on reputational threats and organizational response: silence, denial of allegations, blame avoidance or shifting, and admitting responsibility. As outlined above, how organizations respond to reputational threats depends on the relative threat to their distinct reputation. Existing research shows how certain factors may influence an organization’s decision to opt for a specific response strategy. These conditions for response will be discussed in this section.

The first distinction that can be made is that between silence and talk. Maor, Gilad & Bloom (2013) have developed a set of conditions that influence an organization’s decision to opt for either silence or response. They argue that the strength of the regulator’s distinct reputation shapes its choice between regulatory talk or silence. When the reputation of the organization is relatively strong, the organization can afford to remain silent when faced with reputational threats. On the contrary, when its reputation in a specific domain is relatively weak, or if its reputation is still evolving, organizations may choose a more proactive strategy in responding to external judgements of their performance. They may then choose to either accept responsibility or deny such claims (Gilad, Maor & Bloom, 2013). In other words, the first condition that would shape an organization’s response to a reputational threat is the strength of

(22)

15 the organization’s distinct reputation. If the reputation in a specific area is relatively strong, the organization is more likely to remain silent. In contrast, if the reputation in that specific area is relatively weak, the organization is more likely to respond.

Once an organization decides to respond, it can choose from three different strategies: to deny allegations, to shift the blame to others, or to accept responsibility. Gilad, Maor & Bloom (2013) come up with three considerations that organizations take into account when opting for one of these three strategies. First, considering that an agency will seek to mitigate an immediate threat to its reputation, it is expected that the higher the immediate threat and the greater the risk of political intervention, the more likely it is that an agency will choose to acknowledge the problem and accept its responsibility (Gilad, Maor & Bloom, 2013). Therefore, the second condition that would shape an organization’s response to a reputational threat is the seriousness of the immediate threat. If there is an immediate threat to an organization’s distinct reputation, it is more likely to accept responsibility.

Moreover, when faced with a reputational threat, an organization may seek to mitigate the public debate and controversy over the allegations. Gilad, Maor & Bloom (2013) expect that nonresponse and full admission are less likely to lead to public debate and controversy, whereas blame shifting and denial of the allegations are more likely to stir debate. Therefore, the third condition that may shape an organization’s response to a reputational threat is the explicit wish to close the debate regarding the audiences’ claims. If this is the case, an organization will choose a strategy of nonresponse or accepting responsibility rather than blame shifting or denial of the allegations.

Finally, in choosing a response strategy, an organization will be guided by which message it wants to send about its future practice. Only taking full responsibility will send the message to its external audience that the organization is committed to change its current practice. In contrast, denial of the allegations and blame shifting will consciously send a message that no measures are taken to change its current practice. A strategy of nonresponse does not send a message at all (Gilad, Maor & Bloom, 2013). Therefore, the fourth condition that may shape an organization’s response to a reputational threat is the wish to shape expectations regarding its future practices. If it wants to send a signal that it is committed to change, the organization is more likely to fully admit its responsibility regarding the audiences’ claims.

(23)

16 The various response strategies that organizations may choose from when responding to reputational threats, and their considerations for making that choice have been summarized by Gilad, Maor & Bloom (2013) in the table below.

Table 1: Agency response strategies

Source: Gilad, Maor & Bloom (2013, p. 456)

2.5 Whistleblowing as a reputational threat

Before elaborating on whistleblowing as a reputational threat and how this is likely to affect organizational responses, it would be useful to have clarity on the definition of whistleblowing. Following the definition given by Miceli & Near (1985), whistleblowing is “the disclosure by organization members (former or current) of illegal, immoral, or illegitimate practices under the control of their employers, to persons or organizations that may be able to effect action” (p. 4).

Interestingly, whistleblowing can be seen both as a contribution to an organization’s transparency and accountability efforts, and as a subversion of an organization’s reputation. On the one hand, whistleblowing policies are widely advocated for in order to increase organizational transparency and integrity. As such, whistleblowing can be seen “not as a threat to organizations, but as a necessary part of the regulatory and self-regulatory environment” (Vandekerckhove, Brown & Tsahuridu, 2014, p. 298). Moreover, whistleblower disclosures

(24)

17 could provide useful knowledge on regulatory enforcement (Etienne, 2014). As a result, organizations can use the information provided by whistleblowers to improve their policies, which may ultimately enhance their organizational efficiency (Vandekerckhove, Brown & Tsahuridu, 2014).

On the other hand, whistleblowing can be seen as a threat to an organization’s distinct reputation. An important feature of whistleblowing lies with the intentionality of the action. The whistleblower reports with the expectation that the individual to whom wrongdoing is reported will act on it or publicize it (Johnson & Kraft, 1990, p. 850). Thus, a whistleblower reports the wrongdoing intended to make the information public, with the idea that this will lead to public scrutiny or corrective action. This is in line with the ‘man-made’ character of reputational threats, as described in section 2.3. Moreover, whistleblowing often brings misconduct to light, which may result in heavy criticism to the various dimensions of an organization’s reputation.

Considering that whistleblowing entails the disclosure of “illegal, immoral, or illegitimate practices”, following Miceli & Near’s (1985) definition, whistleblowing may be considered a threat to an agency’s procedural and moral reputation in particular. Allegations concerning illegal practices can be related to the procedural dimension of reputation. When whistleblowers disclose information about an organization not operating according to established rules and procedures, it may harm the procedural reputation of that organization. This can be seen in the cases of Dieselgate and Panama Papers, where the organizations were accused of manipulation of data and tax evasion respectively. On the other hand, allegations concerning immoral practices can be related to the moral dimension of reputation. If an organization fails to be compassionate and to protect the interests of its ‘clients’ and members, whistleblowers may disclose this information and as such harm the organization’s moral reputation. This happened, for example, in the case of Aicha Elbasri, a UN whistleblower who accused the UN’s peacekeeping mission in Darfur of failing to protect citizens by systematically underreporting to the UN headquarters.

The tension between whistleblowing as a contribution to an organization’s effectiveness and whistleblowing as a reputational threat may lead to seemingly irrational choices concerning whistleblower disclosures. Rather than using this information to improve their policies, bureaucracies often dislike to be confronted with their failures and may actively avoid this.

(25)

18 They may show aversion to certain kinds of information or disclosures, even if they could provide useful knowledge on regulatory enforcement and policies (Etienne, 2014). Defensive strategies such as ignoring reports or deploying retaliatory measures are therefore often part of the organizational response to unsolicited disclosures by whistleblowers (Mesmer-Magnus & Viswesvaran, 2005).

Although whistleblowing as a reputational threat has not explicitly been studied before, several bureaucratic reputation scholars have made reference to similar mechanisms. For example, Carpenter & Krause (2012) refer to the ‘bad apple effect’ (p. 29), which occurs when one agent within an organization defines the reputation for that organization. Whereas agencies are often presented as if they were unified actors, the authors recognize that not everyone within the organization may always be on the same page. In that case, an organization’s reputation will be determined by the least common denominator, that is by a reputation-damaging event or by the agent than can be ascribed to it.

In a similar vein, Etienne (2014) has identified employees as a category of third-party informants that may engage in the detection of noncompliant behavior of organizations, “who have been neither solicited by regulators to contribute to detecting incidents, nor have been particularly well received by them when they have done so” (p. 258). Third parties have been relatively understudied as a source of detection information in the existing literature. However, they can pose a significant reputational threat to an organization. Their behavior is unpredictable, their motives are unknown, and they might make disclosures publicly rather than only to the agency. Moreover, if third party informants are at odds with some of the agency’s most important audiences, this may damage the organization’s reputation as well (Etienne, 2014).

Although Etienne (2014) focuses on regulatory agencies, his results can be generalized to other kinds of bureaucratic agencies as well. He argues that detection practices and the use of information by public agencies in general tend to be shaped by reputational considerations. Moreover, his argument that third-party informants, and more specifically whistleblowers, are important contributors to the detection of noncompliant behavior is applicable to other public organizations as well. Unsolicited third-party informants may appear in every type of organization after all. Their contribution is often unpredictable and puts the organization in a rather uncomfortable position, which may significantly harm an organization’s reputation.

(26)

19 Consequently, an organization’s decision to disregard or accept this unsolicited information is largely based on reputational calculations.

In the process of whistleblower disclosures, there are several audiences that the organization needs to manage in terms of their expectations and perceptions, in particular the media, clientele groups, and even their own members. In most cases, the recipient of whistleblower complaints includes the media, which may be able to help the whistleblower generating attention to the perceived wrongdoing (Bosua, Milton, Dreyfus & Lederman, 2014). Often, the media play an important role in bringing the disclosures to light and instigating the debate about who is to be held responsible and accountable. Second, donors or funding agencies are important audiences to manage after whistleblower allegations have become public. Donors may, for example, decide to withhold funding or impose sanctions, either for legitimacy reasons or to exert pressure on the organization to take action. They could also strengthen their monitoring and oversight mechanisms, to ensure that the organization will comply with norms and procedures in the future. Third, whistleblowing disclosures may harm the credibility and trustworthiness of the organization with its clientele groups. For example, allegations of sexual abuse in UN peacekeeping missions has had severe consequences for the trust that local civilians impose in the UN in general. This may have severe consequences for the organization’s ability to carry out its mandate. Fourth, another important audience is the members of the organization itself. After whistleblower allegations, the organization’s reputation may be damaged to such an extent that even its own employees cannot identify themselves with the organization anymore, and they may be incited to leave. The whistleblower himself is often among this audience as well. Either due to retaliatory measures, or due to moral considerations, whistleblowers may decide to leave the organization that they have worked for.

Thus, whistleblowing can be considered a possible threat to an organization’s distinct reputation. Managing expectations of these audiences is of great importance to mitigate negative feedback and prevent further damage to the organization’s reputation. Chapter 5 will apply the theoretical features of reputational threats, as identified in the bureaucratic reputation literature, to the empirical case of whistleblowing within the UN.

(27)

20

2.6 Expectations

Based on the conditions that may shape an organization’s response to reputational threats described in section 3.4.2, we can state several expectations about an organization’s response to whistleblowing as a reputational threat.

Following the line of reasoning of Maor, Gilad & Bloom (2013), if the whistleblower disclosures relate to an area in which the organization’s reputation is relatively weak, the organization is more likely to actively respond rather than remain silent. If the whistleblower disclosure relates to an area in which the organization’s reputation is relatively strong, on the contrary, the organization is more likely to opt for strategic silence.

Further specifying different response strategies, we can expect an organization to choose between denial of allegations, blame shifting, and accepting responsibility. As Gilad, Maor & Bloom (2013) argue, depending on the seriousness of the whistleblower case and the extent to which it poses an immediate threat to the organization’s reputation, the organization will choose a certain response strategy. Assuming that external whistleblowing will always pose an immediate threat to the organization’s reputation, it is expected that the organization will be more likely to admit responsibility as a response.

Gilad, Maor & Bloom (2013) also argue that full admission of responsibility, together with nonresponse, will shorten the life span of the public debate regarding the issue. Assuming that an organization wishes to stifle public debate and controversy, it is expected to choose nonresponse or full admission over blame shifting strategies or denial of the allegations.

Finally, Gilad, Maor & Bloom (2013) argue that the different response strategies may send different signals regarding the change of future practices. Considering the importance for the organization to correctly manage the expectations of its audiences, such as the whistleblower himself and the media, it may want to send a message that it is committed to change future practices that have caused the wrongdoing. Therefore, it is expected that an organization will take full responsibility regarding the audiences’ claims rather than opting for nonresponse, denial, or blame shifting.

(28)

21

3. Research design and methodology

3.1 Research design and approach

In order to find the conditions that shape an organization’s response to whistleblowing as a reputational threat, this thesis has an observational, small-N comparative case study design. Its aim is to study a small number of whistleblowing cases within the context of the United Nations in order to uncover a causal mechanism rather than a general causal effect. To this end, a comparative analysis will be combined with causal process-tracing (Blatter & Haverland, 2012, chap. 2). In doing so, it aims to unravel how the process of organizational responses to whistleblowing unfolds and how this can be explained.

The rationale for a comparative design is that it enables the researcher to find out whether specific events or features produce a certain outcome or not (Blatter & Haverland, 2012, chap. 2). A comparative design allows for comparison between different cases and to compare the variation of certain features of interest with certain outcomes of interest. In order to find the conditions that shape an organization’s response to whistleblowing as a reputational threat, we can evaluate the theoretical expectations across various cases. However, there may be other variables that have an effect on the relationship that is studied. The main strength of the covariational approach, as compared to other research designs, is that these other possible explanations may be controlled for (Blatter & Haverland, 2012, chap. 2). These other explanatory variables are identified in the next section. By keeping these variables as constant as possible across the cases that are studied, it is possible to isolate the effect of the explanatory variable of interest.

The rationale for a small-N study is that it allows for detailed investigation into a specific case. The small-N nature of this thesis allows to take into account contextual factors that may influence the relation that is investigated and to generate thick description (Toshkov, 2016, p. 258). From the literature review, it appears that the conditions that shape an organization’s response to reputational threats are not necessarily straightforward. A small-N study may be helpful to get a better understanding of this relationship. In contrast to large-N studies, small-N studies allow the researcher to take into account specific details. These context-specific details are relatively important to consider in this thesis, as the severity of the whistleblower allegations may differ across cases, as well as their complexity and duration due

(29)

22 to their legal aspects. This may have significant implications for the response strategy that an organization decides to pursue.

3.2 Case selection

The case selection is based on a most-similar systems design, which implies that the cases should vary mainly on the main explanatory variable(s) while the values of all other possible relevant variables remain as constant as possible (Toshkov, 2016, p. 262). The selection of cases that are as similar as possible with respect to the other relevant explanatory variables enables the researcher to ‘single out’ the effect of a particular variable of interest.

The cases have been selected following a two-stage process. First, the research question already delineates the population from which a sample of cases can be drawn. Because this thesis studies cases of whistleblowing within the United Nations, the selected cases are all cases of bureaucratic whistleblowing. This means that they represent the reporting of misconduct in bureaucratic settings rather than in the corporate environment. Moreover, the aim of this thesis is to look at the communication activities of public organizations when faced with whistleblowing as a reputational threat. Therefore, the population from which cases are selected are cases of external whistleblowing, rather than internal whistleblowing. As opposed to internal whistleblowing, external whistleblowing refers to the reporting of wrongdoing outside of the organization. External whistleblowing is more likely to result in negative publicity and may be more harmful to the reputation and image of the organization (Dworkin & Baucus, 1998; De Graaf, 2010). Organizations are therefore more likely to respond to external whistleblowing in terms of their communication activities than they would when facing internal whistleblower disclosures. The decision to consider only cases of external whistleblowing is also based on practical considerations. Cases of internal whistleblowing are much more difficult to identify and select based on certain values of possible explanatory variables. As most organizations that deal with internal whistleblowing cases do not want the information to become public, cases of internal whistleblowing are often surrounded by a high level of secrecy, which makes it more difficult to collect data.

As a second step, cases have been selected considering their values on several possible relevant explanations (control variables) that may influence an organization’s response strategy to whistleblowers. An important control variable is the organizational context, as it may influence

(30)

23 the way in which organizations choose the respond to whistleblower allegations. For example, Miethe & Rothschild (1999) find that more participatory work environments, which are often more cohesive and transparent, demonstrate lower levels of organizational wrongdoing, and therefore also lower levels of external whistleblowing in the first place. Moreover, the role of organizational culture in the organizational response to whistleblowers has been studied within the medical environment by Lachman (2008). She argues that the ethical culture of an organization is an important determinant for their response and management strategy towards whistleblowers.

In order to control for the possibly relevant variable of organizational context, this thesis will focus on cases of whistleblowing within a single organization, namely the United Nations. Martel (2017) points to several organizational features that are specific to the UN context that may explain its relation to its own whistleblowers. First, UN agencies and officials often engage in complex situations that are marked by conflict, human rights violations or other humanitarian issues. As a result, their activities are highly sensitive, which requires a certain level of discretion from UN staff. Second, the organizational context of the UN is marked by its culturally diverse staff. UN staff come from culturally diverse places and operate in various cultural settings, ranging from the UN Secretariat in New York to local peacekeeping missions in (post-)conflict settings. This cultural diversity may have negative consequences for identification and loyalty of staff, however (Saner & Yiu, 2004, p. 247). Third, UN activities are characterized by their geographical dispersion. While the UN Secretariat is based in New York, the UN has offices in other parts of the world as well and many organizational activities take place in environments that are physically remote from the headquarters. By focusing on various cases of whistleblowing within an organization that is characterized by these organizational features, it is attempted to rule out the influence of these different control variables.

Based on these criteria, four cases have been selected that each represent one of the organizational response strategies to reputational threats: silence, denial of allegations, blame shifting, and admitting responsibility. These cases are selected out of a population of eleven high-profile whistleblower cases within the UN. However, the total number of whistleblower cases within the UN is not exactly known, except for the fact that since 2006, 297 requests for protection for retaliation have been filed with the UN Ethics Office (GAP, 2012). The four

(31)

24 cases represent whistleblower cases in different departments of the UN, such as various peacekeeping missions and the Office of the High Commissioner for Human Rights (OHCHR). It should be noted that processes involving whistleblower disclosures tend to be long-running and evolve over time. Therefore, the cases have been categorized based on their characteristics that overlap most with a specific response strategy. The within-case analysis will account for this by analyzing these evolving processes and contextual issues.

A general concern in comparative case study research is selection bias. This may occur when cases are selected based on their values of the dependent variable, which may result in an inaccurate depiction of the causal mechanism. Selection bias is a particularly relevant concern for small-N research designs (Toshkov, 2016, chap. 8). Since the cases for this thesis have been selected based on their values of the dependent variables as well, it is useful to reflect on this. One way to avoid selection bias, while still selecting cases intentionally, is to have a dependent variable that varies (Landman, 2003, chap. 2). In this thesis, the selected cases represent varying dependent variables, namely the different response strategies, which should mitigate the selection bias concern. The cases will be introduced here briefly. In Chapter 4, they will be discussed in more detail.

The cases that have been selected are the following: Aicha Elbasri, Emma Reilly, James Wasserstrom, and Anders Kompass. Aicha Elbasri is a former spokesperson for the UN peacekeeping mission UNAMID in Darfur. In 2013, she blew the whistle on the ‘conspiracy of silence’, referring to UNAMID and UN officials who failed to report wrongdoing committed by the Sudanese government and its militias. Her multiple requests for an independent investigation into her disclosures were in vain, and the UN considered the matter closed without further investigation or comments from its side (GAP, 2014). Therefore, the UN’s response strategy to this case has been categorized as ‘silence’.

The second case is that of Emma Reilly, who is a former official at the Office of the UN High Commissioner for Human Rights (OHCHR) and accused the OHCHR of overstepping its mandate after it had handed over the names of several human rights activists to the Chinese government. The activists were expected to attend a Human Rights Council (HRC) session in Geneva. After her disclosures, the OHCHR strongly rejected the whistleblower’s claims public statement (OHCHR, 2017). Therefore, the UN’s response strategy to this case has been categorized as ‘denial of allegations’.

(32)

25 The third case is that of James Wasserstrom, who is a former official at the UNMIK mission in Kosovo. In 2007, he blew the whistle on alleged kickback schemes involving UN officials and local third parties. Following his disclosures, Wasserstrom faced multiple retaliatory measures and UNMIK used a ‘shoot the messenger method’ to shift the blame to the individual whistleblower, to shift the attention away from the substance of the allegations (Kusari, 2015). Therefore, the UN’s response to this case has been categorized as ‘blame shifting’.

The fourth case is that of Anders Kompass, a former director of field operations for the OHCHR, who blew the whistle on sexual exploitation of children in Central African Republic. Although initially he faced retaliatory measures and was suspended from his office, the UN also acknowledged the problem of sexual exploitation in peacekeeping missions as a result of Kompass’ disclosures and he was exonerated from wrongdoing (Laville, 2016). Therefore, the UN’s response to Kompass has been categorized as ‘admitting responsibility’.

The selected cases, their values for the control variables, and the main explanatory variables are summarized in the covariational table below.

Table 2: Covariational table

Aicha Elbasri Emma Reilly James

Wasserstrom

Anders Kompass Control variable 1

Type of whistleblowing

Bureaucratic Bureaucratic Bureaucratic Bureaucratic

Control variable 2

Venue of whistleblowing

External External External External

Sensitivity of activities

High High High High

Cultural diversity

High High High High

(33)

26 Activities physically remote from UN HQ Independent variable 1 Threat to distinct reputation Independent variable 2 Immediate threat Independent variable 3 Wish to contain or stimulate debate Independent variable 4

Wish to send a message regarding future practices

Dependent variable

Response strategy

Silence Denial of

allegations

Blame shifting Admitting responsibility

3.3 Methodology

Because comparative case study designs cannot generate a deterministic understanding of a causal relationship, this approach is often complemented with within-case analysis (Blatter & Haverland, 2012, chap. 2). The covariational approach is well-suited to establish a correlation between the independent and dependent variables, but evidence from causal process-tracing is necessary to substantiate this claim and to establish a causal mechanism. Within-case analysis by means of causal process-tracing strengthens the internal validity of this thesis, and makes it possible to strengthen the claim that the explanatory variable under investigation really produces the outcome that we are interested in (Blatter & Haverland, 2012, chap. 3).

Causal process-tracing is especially useful when the similarity of the cases, which is the main premise of comparative analysis, cannot be completely warranted or when the covariational analysis is indeterminate, which is the case when more than one explanatory variable may cause the outcome variable of interest (Blatter & Haverland, 2012, chap. 3). While the cases for this research have been strategically selected, they may still vary on other possibly relevant

The effect of whistleblowing as a reputational threat on the United Nations’ communication strategy