A Comparative Study on EU Regulatory Agencies: Institutional Isomorphism in the midst of Bureaucratic Reputation Uniqueness

(1)

6/11/2019

A Comparative

Study on EU

Regulatory

Agencies:

Institutional Isomorphism in the

midst of Bureaucratic Reputation

Uniqueness

STUDENT: Oladeji Akintunde

S2109050

SUPERVISOR: DR. DOVILĖ RIMKUTĖ 2ND READER: DR. MADALINA BUSUIOC

MASTER THESIS PUBLIC ADMINISTRATION

(2)

pg. 1

1. INTRODUCTION

1.1 Looking beyond EU-agency uniqueness

With the rise of EU-regulatory governance ((Busuioc, 2009; Carpenter, 2001, Groenleer, 2009; Majone, 1997), agencies continue to compete for resources, reputation, legitimacy, amongst other aspects by differentiating themselves from others. Their main goals are to convey their competence and ultimately survive. However, as rational entities, one is poised to ask if agencies are really different when it comes to strategically achieving their goals under similar constraints. In looking at a significant reputation-focused study conducted by Busuioc and Rimkute (2019), their goal was to highlight how EU regulatory agencies (ErAs) legitimize their existence with regards to reputation management -because according to Carpenter (2010), there are several ways in which a public institution can communicate its reputation to its audience- by way through a study on their behavioural practices. The authors’ results have carved out an intriguing paradoxical puzzle. This paradox simply conveys a trend whereby ErAs generally depict strategic similarity (Wæraas & Sataøen, 2014) as they appear to focus on either communicating the technical –which according to the European Commission is their raison d’etre (Busuioc & Rimkute, 2019; European Commission, 2002; Groenleer, 2009) or performative aspect of their reputation despite the theoretical justification that agencies aim to convey their unique reputation (Carpenter, 2010; Carpenter & Krause, 2012; Wæraas & Maor, 2015) through differentiation (Wæraas & Sataøen, 2014).

Expatiating more on this puzzle, reputation management entails communicating either technical, performative, procedural or moral reputation (in no particular order) (Carpenter, 2010). Moreover, while these tenets are theoretically separate, they may not be empirically distinguishable (Carpenter, 2010; Carpenter & Krause, 2012). Several studies (Bankins & Waterhouse, 2018; Wæraas & Sataøen, 2014) point towards the fact that agency differentiation can be beneficial in terms of being able to set themselves apart from others who are not as ‘competent as they are’ as in the case of a businessman putting a little spin in marketing a product that is otherwise sold by some 50 more business men and women. However, equally important is the need to be similar to others alike in the same environment in order to be patronized. An example of this is with regards to hospitals that have to convey that they meet certain standards in a bid to be patronized, subsequently engaging in convergence (Deephouse & Carter, 2005). This conveys the need to create a balance between differentiation and similarity (Bankins & Waterhouse, 2018; Beckert, 2010; Deephouse & Carter, 2005;

(5)

pg. 4 Deephouse & Suchman, 2008; Wæraas & Sataøen, 2014). Nevertheless, when it comes to EU agencies, technical reputation management strategy has remained fairly steady amongst EU agencies and performative reputation appears to be growing within the regulatory environment over time, relative to the other reputation dimensions (Busuioc & Rimkute, 2019).

While this view can be justified by their functions (Groenleer, 2009; Busuioc & Rimkute, 2019), not much is known about how ErAs actually work in practice (Groenleer, 2009). Moreover, the varying ambiguous mandates that fail to depict the entirety of an agency’s task (Groenleer, 2009), coupled with the number of audiences under each EU agency’s radar (Carpenter, 2001, 2010; Carpenter & Krause, 2012) has done little to supplement the justification for reputation management similarities. As a result of this, it is difficult to dwell solely on such functional explanation. Consequently, an alternative view can be derived from institutional isomorphism wherein the focus is on highlighting the key mechanisms resulting in institutions appearing similar, be it in regards to functions, strategy or a general organizational outlook (DiMaggio & Powell, 1983). In this case, the focus is on reputation communication strategy. In addition, because isomorphism entails a potential domino effect vis-à-vis the institutional impact, there is an understanding that this puzzle can be explained with a consideration of time factor.

Moreover, as with any group within the same environment, there is the tendency to convey that what other competitors are doing can be replicated and done better. However, to compete with others at what you do, there is a need to go beyond copying to be better (Beckert, 2010). Taking a different perspective, copying may be the solution for institutions in times of difficulty and pressure wherein agencies are either pulled or pushed towards the rational solution. External audiences can exercise constraints on agencies and pushed them towards a certain solution or agencies may be pulled to a common solution in view of internal uncertainty. Moreover, the push or pull factor can be more pertinent depending on the age of the agencies. This has some implications, as path dependency (Pierson, 2000) can be seen as a key marker to determine what trajectory is followed, the likelihood that the solution may not be copied correctly, and the fact that the solution may have been applied in a different context.

1.2 About this study: Research Question

In light of this conundrum, despite the need for reputation differentiation, the question under focus is:

(6)

pg. 5

Are agencies actually copying each other, if so, what are the likely pathways leading agencies to appear similar despite their push for a unique reputation?

To answer this question, I attempt to explain organizational behaviour with regards by complimenting it with isomorphic tendencies. As such, this interplay will direct the focus of this study.

Cultivating reputation is an important goal of ErAs as it cements agencies survival and aid in cutting across other critical aspect of its functions (Carpenter, 2001; Busuioc & Lodge, 2016). Fostering reputation through an efficient delivery on mandated tasks is difficult to attain, let alone maintaining this reputation (Eccles, Newquist, & Schatz, 2007). Moreover, the relativity of reputation within an environment makes the framing audience’s perception the cost-effective option. This is because perception can sometimes trump reality especially in situations where evidence cannot easily be provided (Oliver, 1991). That said, the effect of pulling this off is highly desired, however not all agencies are capable of it. As a result, agencies come under varying pressure at different periods in their organizational lifespan, from different multifaceted audiences (Busuioc, 2009; Carpenter, 2001, 2010; Carpenter & Krause, 2012; Wæraas & Maor, 2015). This constant interplay between organizational pressure and agency rationality leading to isomorphism has ultimately increased the development of a dominant reputation strategy.

Moreover, the literature on bureaucratic reputation has recently attributed the build-up of the communication of technical reputation over time as one concerned with agency’s overall survival (Rimkute, forthcoming). Perhaps this can explain the pattern of EU agency’s communication of technical reputation or the mere fact the agency’s focus in technicality is the “raison d’etre” (European Commission, 2002, p.5). Nonetheless, the prioritization of performative reputation over other aspects remains a key area of clarification.

In demonstrating this phenomenon, the use of crisp-set qualitative comparative analysis (csQCA) (Ragin, 1987) provides rich methodological power in unveiling if indeed agencies do engage in imitative behaviour in light of pressures of different calibre. The method also enables an explanation for the drive towards performative and technical reputation. Indeed, the result conveyed, some aspects of pressure do serve as key determinants in triggering imitative behaviour when considering reputation management, though other forms of pressure remains speculative.

(7)

pg. 6 1.3 Relevance of the Research

Looking at the relevance of this study, it makes a stark connection between reputational and institutional studies. Specifically, the prime purpose here is an attempt to explain the trend within EU regulatory governance wherein ErAs tend to generally communicate their technical and performative reputation in spite of the push for uniqueness.

While this study attempts to combine two theories in a complimentary manner, I do not aim to provide all the details regarding these theories. The relevance of reputation management with regards to varying dimensions of isomorphic pressures is the bone of contention. For the most part, this research focuses on ErAs –the unit of analysis– which is not to say the findings may not be relevant for other population of cases. In addition, addressing the impact of agency’s passive and activeness, and the impact of time factor in responding to isomorphic pressures are areas this thesis aims to shed light on. The goal of this research is therefore to venture into a new theoretical understanding of agency behaviour.

1.4 Structure of this thesis

To delve into this study, the next chapter will address important works in the institutional and organizational reputation scholarship and the impact on the focus of this research. What follows in chapter 3 is a theoretical framework highlighting the mechanisms at play; explaining the interaction of the conditions is key. More so, the outcome in line with the causal conditions will be discussed. Through this discussion, expectations will be subsequently posited.

Chapter 4 will put forward the research design that will be taken to test the claims within this study in line with the theoretical underpinning. The benefits of using csQCA for this study will be outlined. Thereafter, it would be crucial to present and offer a clear understanding of how the concepts will be operationalized measure.

Having done the aforementioned, in line with csQCA, what follows in chapter 5 and 6, the results will be presented and subsequently discussed with reference the theoretical framework and empirical examples. Furthermore, I will address the limitations of this study. Last but not least, in chapter 7, I will provide a surmised summary of this study, I then address the reliability and validity of the study before offering possible subject areas for further studies.

(8)

pg. 7

2. LITERATURE REVIEW

In this chapter, a recollection of the organizational and institutional study will be carried out. Moreover, it would be important to address what previous studies have done with regards to attempting in varying degree to conflate reputation-seeking organizations and isomorphic tendencies. But first, it is important to set the scene.

Organizational study has been a topic of discussion amongst scholars of diverse fields for a long period of time (Busuioc, 2009; Keleman 2002; Jennings, 2010; Carpenter, 2001; Wæraas, & Maor, 2015). The significance of these development offers the possibility of comprehending good governance (Moffitt, 2010). More importantly, in the present layout of governance, there are no hiccups as to the growing need to uncover the institutional workings to get a better grasp on what is unfolding with regards to supranational entities like the EU and public administration (Groenleer, 2009). There has been a push to uncover the behavioural make-up of institutions -public and private- which has led to a focus on reputation (Bankins & Waterhouse, 2018; Carpenter, 2001; Deephouse & Carter, 2005; Maor, 2015). Furthermore, reputation becomes a key tenet of organizational study because of the impact it offers in shaping an organization’s image and identity (Carpenter, 2010; Carpenter & Krause, 2012; Bankins & Waterhouse, 2018) as well as its impact on autonomy and legitimacy (Busuioc, 2016). Crucially, scholars such as Carpenter & Krause (2012), Maor (2013, 2015), and Wæraas & Byrkjeflot (2012) have created a broad path to walk on with regards to answering these questions. That said, it is important to briefly discuss EU-agencies.

2.1 What drives EU agencies?

The emergence of decentralized agencies (also ErAs) in Europe has not been without its issues (Groenleer, 2009). For example, the issue of agency accountability and control has been amongst the growing area of debate (Busuioc, 2009; Vos, 2005). In spite of this, ErAs have grown to become part and parcel of EU governance (Dehousse, 2008; Egeberg & Trondal, 2017). In reality, ErAs are black-boxes; what is known about them is that they are legal rational entities with varying degree of formal autonomy (quasi-autonomous) tasked with regulatory, monitoring and coordinating abilities (Groenleer, 2009). Put simply, they have an “important role in implementing EU legislation and in regulating European policy sectors” spanning social, political and economic facets (Groenleer, 2009, p.15). Moreover, the creation of the

(9)

pg. 8 ErAs, thought to be one of the “most prominent institutional innovation” and a “genuine process of agencification” is expected to bring Europe closer to its citizens (Groenleer, 2009, p.15). In fact the rise in the number of ErAs is unprecedented (Davies, 2013). The conventional principal-agent theory has received much criticism over the years (Dehousse, 2008) because ErAs as rational entities are self-aware and have the main goal of organizational survival. Moreover, resource limitation has made the EU-regulatory competition fiercer (Groenleer, 2009) especially in the face of pressure to improve effectiveness and efficiency (Borrás et al., 2007; Groenleer, 2009). Nonetheless, these pressures entails agencies need to employ strategic choices (Palmer, Jennings & Zhou, 1993).

In view of the aforementioned, does agency rationality imply that ErAs will be similar when it comes to communicating their reputation strategy? To provide a possible answer, it is crucial to take a retrospective view of EU agencies. Being the first two ErAs, Cedefop and Eurofound were very much focused on delivering technical expertise as opposed to an alternative focus i.e. performance, procedure (Cedefop Press Release, 2015). Their decision to maintain the same strategic focus sheds some light on the rationale of agencies in general. It can be expected that rational agencies would use Eurofound and Cedefop as benchmarks for regulatory success; this further extends to how they manage and what reputation strategy they communicate to their audience. Consequently, institutional theory helps to shed light on the relation between agency rationality and copying, as such, an overview of the literature on institutional and reputational studies will follow.

2.2 Institutionalization as a theory

The literature on institutionalism has been extensively dealt with in many respect (see Bevir, 2009 and Greenwood, Oliver, Suddaby, & Sahlin, 2008). Moreover, neo-institutionalism as an encompassing umbrella of institutional theory has come a long way from its presentation over four decades ago by Meyer and colleagues (Meyer & Rowan, 1977; Scott & Meyer, 1983), as well as DiMaggio and Powell (1983). One influential area of focus is on sociological institutionalism (SI) (Beckert, 2010). SI focuses on the varying means in which institutions come to manifest their potential capabilities and the forces that guide their actions or lack thereof. The general understanding categorizes institutions as regulatory structures, governmental agencies, laws, courts, and profession (Oliver, 1991). The fundamentals of the institutional study has conflated different theories stemming from strategic choice (Oliver, 1991), resource dependence (Pfeffer and Salancik, 1978), ecology and politics (Palmer et al.,

(10)

pg. 9 1993) amongst others. Nonetheless, organizations are thought to strive for relevance and legitimacy, in hopes of survival (Palmer et al., 1993). In so doing, these processes can “make organizations similar without necessarily making them efficient” (DiMaggio & Powell, 1983, p.147). SI still has more to offer. One particular area of interest refers to the implicit and explicit foci of multi-level organizational cooperation by way through a push or pulling factor (Beckert, 2010). In view of the aforementioned, institutional isomorphism (DiMaggio & Powell, 1983) captures much of the discourse on neo-institutional study, but what does isomorphism entail? 2.2.1 What is Institutional Isomorphism?

DiMaggio and Powell (D&P) and Meyer and Rowan’s studies have formulated the study on institutional isomorphism. DP articulated three distinct yet interrelated concepts circumventing isomorphism. Prior to that, Meyer and Rowan’s formulation through their study conveyed the impact of “the rationalization of interaction and professionalization of elites in a field” (Palmer et al., 1993, p.126) wherein organizational structures come to “reflect rules institutionalized and legitimated by and within the state” (DiMaggio & Powell, 1983, p.150) even when based on myths about organizational behaviour (1977). In that respect, D&P have argued that over time, institutions within a socially constructed environment come to resemble each other as a result of the institutional pressures from internal and external forces (DiMaggio & Powell, 1983; Pache & Santos, 2010). These convergence can be in parts through structures, processes, strategies etc. Moreover, these changes can work its way into changing the complete outlook of the organizations and lead to organizational similarity (DiMaggio & Powell, 1983). According to D&P, isomorphism is made manifest through three mechanism namely coercive, mimetic and normative. [1] Coercive isomorphism centres on the influence of power and dependence (Pfeffer and Salancik, 1978). It is understood to be driven by two forces; pressure from other institutions which can determine the flow of resources and pressure to deal with the “cultural expectation of the larger society” (Mizruichi & Fein, 1999, p.657). The former relates to the dependence theory wherein organizations need to ‘behave themselves’ with regards to certain powerful constituents because they have control over the flow of resources. The latter takes into account the impact of legality on organizational action by acting as a constraint. [2]

Normative isomorphism is represented through professionalization or as termed by Oliver

(1991), attractiveness. This refers to the socialization of those within the varying organizational field which suggests that a force pulls or pushes organizations into appearing similar. In short, the broad internalization of norms closely explains how normative isomorphism functions as a form of external and internal pressure (Frumkin & Galaskiewicz, 2004). This further touches

(11)

pg. 10 on Mizruichi & Fein’s study which makes mention the impact of social pressures by members of other organizations because of the influences within social networks (1999).

Perhaps one of the most defining aspect of D&P’s study is the third mechanism, [3] mimetic

isomorphism (Mizruichi & Fein, 1999). This form of pressure centres on the notion that

institutions are pulled towards a certain strategy (that has been successfully tested) in the wake of organizational uncertainty (DiMaggio & Powell, 1983). Like normative isomorphism, the system in which institutions model themselves after would tend to have a drawing or attractiveness effect, offering itself as a point of reference (Oliver, 1991).

There has been numerous attempts to reconceptualise these mechanism or take a different approach (See: Mizruichi & Fein, 1999). Reasons stems from the complex nature of D&P’s isomorphic pressure, especially when acknowledging the difficulty with empirically differentiating the manifestation of these isomorphic tendencies and their ensuing effects or fully capturing the dynamics of pressures (Boxenbaum & Jonsson, 2008; Mizruchi & Fein, 1999). Others have combined the strengths of institutional studies with alternative views in a complimentary fashion (Galaskiewicz and Wasserman, 1989; Haveman, 1993; Han, 1994). This further conveys D&P’s work in offering itself as a broad ranging supplementation to organizational studies. To capture this dynamic formulation, the literature has associated “increased prestige, stability, legitimacy, social support, internal and external commitment, access to resources, attraction of personnel, fit into administrative categories, acceptance in professions, and invulnerability to questioning” (Oliver, 1991, p.150) with organizational conformity.

In this vein, I have chosen to provide my own rendition of D&P’s mechanisms which has similar bearings theoretically but at the same time can be empirically measured to determine the effects of certain pressures on the decision to communicate a certain reputation strategy. These concepts are initial, internal and external pressure. Moreover, as depicted by Mizruchi & Fein, taking such trajectory has implications (1999).

2.3 Implications of Institutional Pressures

When it comes to the impact of institutional pressure, the general understanding is that it could lead to convergence (DiMaggio & Powell, 1983). This is the case for institutions like ErAs, however the impact of time in the ensuing outcome is an important requirement. Agencies within the EU are structurally and functionally different at the stage of establishment

(12)

pg. 11 (Groenleer, 2009). For example, looking at the purpose for the creation of ErAs, the first 2 were tasked with a view to producing and spreading information pertaining to EU interest while the next batch were mainly tasked with being instruments for implementing EU policies (Inter-institutional Working Group on regulatory agencies, 2010a). What is more, post-2000 saw ErAs created after that period tasked with “providing independent scientific/ technical advice/ information” and fostering cooperation (Inter-institutional Working Group on regulatory agencies, 2010a). The varying mandates of these agencies plays a part in the creation of a heterogeneous cluster of agencies. However, in due time and in consideration of the combination of pressures, agencies can come to look similar in various aspect (DiMaggio & Powell, 1983). In other words, there is either a reinforcing mechanism driving agencies towards homogeneity or agencies themselves engage in this isomorphic acts.

Two crucial inferences can be made from these views. The first comes from the ability of agencies to be actively or passively pulled or pushed towards a strategic path of copying. The second aspect refers to the fact that rationality is a common feature of ErAs as their decisions are carefully thought through in the midst of bounded rationality (Burns & Wholey, 1993). Hence, there appears to be a production of a dominant solution, seeing as what is important for them is survival.

Furthermore, the literature on institutional isomorphism has generally taken the idea of agencies copying each other to be the dominant view (DiMaggio & Powell, 1983; Beckert, 2010; Levitt & Nass, 1989), however what has not been done is to differentiate adaptation from copying (Frumkin & Galaskiewicz, 2004). By looking at institutional isomorphism through the lens of adaptation, as opposed to copying, theoretical clarity can be attained. Agencies copying each other usually denotes passivity, but the literature implies otherwise as an agency’s response to pressure can differ from passive to active (Oliver, 1991). Moreover, copying entails a word for word type of approach however, this is problematic because with bounded rationality, the chance of fully replicating a model can be difficult. In addition, adaptation is preferred as a useful conceptual tool because while it entails some aspects of copying, it functions as a key marker for the consideration of ultimately conveying the ability to explain an agency’s activeness in the sense of being able to act rationally. A simple illustration can be found in figure 1. That said, both terms will be interchangeably utilized throughout this study.

(13)

pg. 12

2.4 The power and impact of reputation management in defining agency future

The literature on neo-institutionalism has for the most part dealt with the impact of conformity on legitimacy (DiMaggio & Powell, 1983; Meyer & Rowan, 1977; Heugens & Lander, 2007; Oliver, 1991; Ashforth & Gibbs, 1990) but what is not taken into full consideration is the impact of reputation on convergence (Deephouse & Carter, 2005; Wæraas & Sataøen, 2014). For example, the impact on focusing on legitimacy building can be costly (Mizruchi & Fein, 1999) for varying agencies because of the link to performance. Building a reputation is an effective and cost-efficient way for agencies to function (Deephouse & Carter, 2005). Moreover, the focus on reputation building opens more doors to institutional goals such as becoming more autonomous, legitimate, having an influence in governance (Maor et al., 2012; Nicholson-Crotty & Miller, 2012). As a result, it can lead to survival (Carpenter, 2001; 2010). This can be the most preferred for some agencies especially when the resources available to them are limited.

The literature on reputational study has provided minimal coverage with regards to agencies copying each other, however, Carpenter does makes mention of the impact of dealing with threat which can affect the way competitors within similar field respond (2010). Furthermore, Wæraas & Sataøen come close in their (2014) study as they convey that “reputation management literatures may be overstating the importance of differentiation, while ignoring the benefits of similarity” (p.251). Indeed, “differentiation may not be an ideal solution even if a field is characterized by growing competitive pressures” (Wæraas & Sataøen, 2014, p.251). Wæraas & Sataøen further depict hospitals as being in a highly institutionalized organizational field (2014); ErAs are no different vis-à-vis being in a strongly institutionalized environment. This conveys the presence of strong professional norms which can consequently lead to isomorphism (DiMaggio & Powell, 1983). In view of the aforementioned, there is a need to uncover the dynamics of organizational study and its impact for agency behaviour.

COPY ADAPT

(14)

pg. 13 2.5 Organizational Perspective

Organizational reputation at the public sector level has only recently acquired necessary attention amongst varying disciplines (Carpenter, 2001, 2010; Wæraas & Maor, 2015). Carpenter (2010) capitalized on the regulation and organizational theories focusing broadly on the rational and psychological nature of institutions. This in effect has brought about the rise of bureaucratic reputational scholars (Busuioc, 2016; Carpenter & Krause, 2012; Gilad, 2015; Maor, 2012; Rimkute, 2018; Wæraas & Byrkjeflot, 2012) who have given much attention to the impact of reputation in organizational behaviour.

According to Carpenter, organizational reputation is “a set symbolic beliefs about the unique or separable capacities, roles, and obligations of an organization, where these beliefs are embedded in audience networks” (2010, p.45). This creates a connection to an organization’s identity (Gilad, 2015) which has to be inbuilt within the respective institutions for them to actually believe in the ‘creed’. In other words, organizational identity can shape reputational concerns and vice-versa through internal and external factors (Bankins & Waterhouse, 2018; Gilad, 2015; Gilad & Yogev, 2012; Wæraas & Byrkjeflot, 2012). In addition, identity need not be conflated with an organizational image (Bankins & Waterhouse, 2018) because in the advent of reputation management, what an organization conveys might be different from what they actually do (ICF Consulting Limited, 2017) and that to a certain degree is due to survival instincts. Reality after all, can be different from perception (Durrheim, 1997).

Bureaucratic reputation scholars further classify identity to represent something that is created within an organization, while image represents external views about a given institution (Carpenter, 2010; Bankins & Waterhouse, 2018). The understanding is that external image can affect how the identity develops. This has an implication on how agencies will be viewed because the push for unique reputation can be hampered by external pressures (Carpenter, 2001, Carpenter & Krause, 2012; Maor, 2013; Maor & Sulitzeanu-Kenan, 2015). Consequently, if a message is successfully communicated to the right audience, it could align organizational identity with image which can in effect be necessary for communicating a strong reputation (Wæraas & Byrkjeflot, 2012; Wæraas & Sataøen, 2014).

With respect to the reputation strategy communicated, there are four dimensions to focus on, namely technical, performative, procedural and moral reputation (Carpenter, 2010; Carpenter & Krause, 2012; Wæraas & Maor, 2015). In practice these dimensions are intertwined, nonetheless it is somewhat easier to distinguish between the dimensions based on regulatory

(15)

pg. 14 tasks. For example, the European Banking Authority (EBA) is likely to focus on dealing with the technical reputation (Everson, 2012). But then again, the picture painted is not so clear cut because depending on the impact of its multiple overlapping audiences and its various tasks and organisational goals (Gilad, 2015), the dimensional focus may be altered frequently. As Carpenter surmised that “when trying to account for a regulator’s behavior, look at the

audience, and look at the threats” (2010, p. 832; italics in original). Agencies are always under

varying degree of pressure by a network of audience, consequently, they constantly have to communicate the reputation (Carpenter, 2010).

Furthermore, as highlighted above, discussing reputation management requires addressing reputational threats and the strategies available to agencies (Deephouse & Carter, 2005; Gilad & Yogev 2012; Maor, Gilad, & Bloom, 2012). As insisted by Maor and Sulitzeanu-Kenan, “the notion of reputational risk is one of the cornerstones of reputation-driven conceptual framework” (p.33, 2015). Mishandling threats can be consequential for reputation management however, depending on agencies’ abilities, they are able to exercise strategic silence (Maor et al., 2012) which can enable them avoid blame from regulatory failure (Moffitt, 2010; Maor & Sulitzeanu-Kenan, 2013; Hood, 2002, 2011). In addition, akin to Meyer & Rowan (1977)’s view on organizational behaviour, agencies can engage in regulatory talk (Maor, 2013) to depict their capabilities (Carpenter, 2010; Gilad et al., 2013). What is more, agencies can also employ risk aversive strategies (Hood, 2007) in order to insulate themselves from harsh and reputation threatening issues. However, it is noteworthy to mention that not all organizations can effectively capitalize of communicating its reputation.

Additionally, achieving a unique reputation (Carpenter 2010) can be difficult, not to mention maintaining that reputation. However, it has generally been taken by the organizational scholarship to be an automatic switch wherein unique reputation can be acquired albeit under pressing conditions (Maor & Sulitzeanu-Kenan, 2015). What has been somewhat neglected is what happens to new agencies (Maor, 2015); do they have to shadow other organizational bodies? (Maor, 2013, Carpenter & Krause, 2012), what gives them the confidence that their work will be widely acknowledged as important, how much work would new agencies need to carry out to expand, maintain and preserve the level of reputation that is attributed to each of them or as discovered by Luoma-Aho (2007), is it indeed better to aim for neutral as opposed to an excellent reputation. These questions cannot easily be answered due to the contextual nature, however, that should not hinder an attempt to seek clarity.

(16)

pg. 15 2.6 The Interaction between Institutional Isomorphism and Bureaucratic Reputation

Till this day, there has been little to no study on the impact of reputation management in pulling or pushing agencies –ErAs in particularly- towards convergence (Frumkin & Galaskiewicz, 2004; Deephouse & Carter, 2005; Wæraas & Sataøen, 2014). As such this study offers a fairly nuanced outlook. Moreover, Maor and Sulitzeanu-Kenan have also attest to the fact that several studies have “devoted limited attention to the impact of reputational consideration” (2015, p.31). According to findings from Busuioc and Rimkute (2019), performance and technical aspects has been the reputational focus for ErAs over time. In view of this, proper management of bureaucratic reputation can yield beneficial results, amongst which can be considered a part of organizational goals. However, not all institution are able to effectively manage its reputation as it is not always clear-cut the route to take amidst several underlying issues. Consequently, institutions may be pulled or pushed either by external or internal factors across different period (Bankins & Waterhouse, 2018; Wæraas & Sataøen, 2014) -with a keen focus on varying constituents- towards models that appear to be dominant within an environment.

(17)

pg. 16

3. THEORETICAL FRAMEWORK

This section puts forward several propositions in a bid to address the research question: what

are the pathways leading ErAs to appear similar in their reputation communication? In light

of this, an explanation as to the presence of dominant reputation -technical and performative- and the choice to communicate either of these reputation dimensions thereby leading to convergence will be posited. Owing to the novelty of this research, I have used informed deduction though concepts have been extracted from varying aspects of the literature in a bid to create an integrative explanation through the theories in application. Additionally, the method to be discussed in the next chapter will justify the decision to choose the theoretical trajectory and the research question as a whole.

3.1 Outcome: Communicating a dominant reputation

Actual performance it appears is not the only means by which to gain accreditation as ErAs are able to cultivate for themselves a reputation that is worth taking seriously (Carpenter, 2010). Reputation is a “distinctive characteristic of an organization that differentiates it from other similar organizations in the polity and emphasizes the agency’s exclusive character and unique regulatory activities” (Rimkute, 2018, p.73). Moreover, building a reputation entails the consideration of internal and external views of an organization (Bankins and Waterhouse, 2018). In other words, it aids in the reduction of uncertainty with regards to competency (Lang, 2015, p.1). In that respect, it goes without saying that a successfully managed reputation can provide legitimacy, autonomy, discretionary power amongst others, in order to ensure a smooth sailing organizational function (Bertelli, Sinclair & Lee, 2015; Busuioc, 2009; Carpenter & Krause, 2012; Groenleer, 2009).

While regulation governance can be complex especially when considering the case of the EU, owing to the number of actors involved in the ‘agencification of Europe’ and the possible stringent ex-ante control (Levi-Faur, 2011; Egeberg & Trondal, 2017; Wonka & Rittberger, 2010; Busuioc, 2009), the politicization of the role of regulatory agencies has created the need for survival through reputation management vis-à-vis those who can be classified as regulatory constituents (Busuioc & Lodge, 2016).

Bureaucratic reputation entails addressing four dimensions that though in theory appear independent, in principle are somewhat intertwined and can overlap (Carpenter, 2010; Busuioc & Rimkute, 2019). The dimensions refer to a focus on moral, procedural, technical and

(18)

pg. 17 performative reputation (Carpenter, 2010; Carpenter & Krause, 2012; Busuioc & Lodge, 2016). Moral reputation entails the capability of agency to portray itself as selfless to the point where those within and outside the agency are prioritized (Carpenter & Krause, 2012). Procedural reputation as conveyed by Carpenter and Krause, “does the agency follow normally accepted rules and norms, however good or bad its decision?” (2012, p.27). So in that sense, agencies’ attempt to portray themselves as rule followers bodes well into the procedural reputational claims. With technical and performative reputation -the two most important area according to Busuioc & Rimkute (2019) - the former appears to look at the tentative ability of agency to have the necessary know-how and know-why for dealing with potential complex regulatory issue area irrespective of a consideration of its actual performance. The latter refers to the ability of agency to carry out its task efficiently and competently (Carpenter & Krause, 2012). That said, it is generally the desire for regulators to have control of these dimensions and attend to each of them but in reality, trade-offs have to be made especially with regards to those classified as its audience (Busuioc & Rimkute, 2019; Busuioc & Lodge, 2016). Based on figure 2, the trend conveys that over time, two general dominant reputation strategy have grown at different paces, with performative reputation becoming most dominant in 2015. Therefore, due to this invariable importance, this study focuses on the two most recurring reputation dimensions. 0 2 4 6 8 10 12 14

PERFORMATIVE REPUTATION TECHNICAL REPUTATION

Figure 2 Dominant Reputation Strategy Score for EU Regulatory Agencies (1976-2017) extracted from Busuioc & Rimkute (2019)

(19)

pg. 18 For the most part, technical expertise appears to be the most referenced when considering the EC’s operating framework for ErAs; it goes on to further convey that the basis of operation lies in their “technical and/or scientific assessment” (European Commission, 2002). Even with that, the data in figure 2 conveys the normalization of technical reputation over time, offering another view into institutional isomorphism. Nonetheless, with the trend clearly portraying a rise in performative dimension, despite the steady technical base (Busuioc & Rimkute, 2019), it would be interesting to pit the two dimensions against each over time to uncover what mechanism in involved in the dynamic change.

Busuioc and Rimkute (2019) claim that technical reputation is prioritised before ErAs develop the need to look to other dimensions owing to the different audiences and their diverging interests. While this is highly plausible, the main focus on performative dimension as an alternative appears to still point to the fact agencies converge when it comes to reputation strategy. Moreover, if each regulatory agency claim to have a unique identity which in principle means they are more or less mixing and matching between the reputational dimensions (Deephouse & Carter, 2005), the trend towards a similar reputational claim should not be present, which clearly is the case (Busuioc & Rimkute, 2019).

A plausible reason is that agencies in principle aim to be reputable on all dimensions (Carpenter, 2010) but are unable to do this due to varying reasons. Consequently, the next best thing may be to focus on multiple but not all dimensions. Even with that, it does not necessarily explain the trend. Again, another explanation for agency’s focus on performative reputation can be inferred from Busuioc and Rimkute, as they highlight that “technical goes hand-in-hand with performative within the EU regulatory state and, thus, it makes sense (...) that the emphasis on the technical dimension would over time, broaden out to an emphasis on performative aspects” (2019, p.20). Put simply, performance and technical expertise can be seen as the crux of an agency’s existence. In addition, as argued elsewhere, ErAs attempt to focus on these two dimensions to improve their reputation (Rimkute, 2018; Busuioc & Rimkute, 2019; Groenleer, 2009; Groenleer, 2014). “Unique reputation” (Carpenter, 2010) however would only function if regulatory agencies are pursuing a certain strategy that does not reflect what others are also doing. Nonetheless, there appears to be conditions that constrains organizational behaviour and consequently encourages homogeneity across field over time.

(20)

pg. 19 3.2 Conditions Leading To Isomorphic Outcomes

In the review of the literature, a conflation of concepts was posited that shed light on the strategic response of an organization to pressure while paying close attention to the isomorphic tenets of DiMaggio and Powell (1983). However, because of the focus on homogenization as opposed to divergence, there is a specific selection of conditions. Furthermore, the novelty of this study deems it important to develop the concepts so as to provide a base for further analysis and meaningful hypotheses creation. It is therefore argued that pressures of varying forms can lead to convergence in reputation management (DiMaggio & Powell, 1983; Meyer & Rowan, 1977). These pressures can be categorized into three: [1] Initial-pressure: this considers the innate pressures that organizations deal with upon its start-up phase. Pressures in this respect can arise from being a relatively new agency in the environmental field, the level of formal independence level, and the subsequent policy saliency level upon creation. [2] Internal pressure: this refers to the pressure agencies feel in order to instigate changes, and this could result from uncertainty which is similar to the concept of mimetic isomorphism (DiMaggio & Powell, 1983; Pache & Santos, 2010). [3] External pressure: the final form stems from the pressure exerted by constituents (Carpenter, 2010) outside the agencies in a bid to instigate a preferred response.

It is clear that the role of a regulatory agency’s audience carries weight (Carpenter, 2010; Wæraas & Sataøen, 2014), as such, pleasing multiple audiences is not an easy task (Bususioc & Rimkute, 2019) especially when considering the fact that reputation is multifaceted (Gilad et al., 2013). Considering the relative reputation of the agency and the principal being at stake (Hood, 2007), it comes as no surprise the scale of pressure that is exerted at varying levels. Even despite the fact that agencies tend to make concessions considering their audience, Wæraas and Byrkjeflot made it clear that sometimes principals can throw agencies under the bus for the sake of their own political acquisition goal (2012). Moreover, cooperation with agencies suffering from reputation loss can be detrimental. That said, it is crucial to look more into these pressures.

INITIAL-PRESSURE

It is fair to state that agencies face different form of pressures, and its inception period is no exception (Groenleer, 2009). How an agency kicks off its functional campaign can potentially determine the level of reputation or momentum they can cultivate with regards to their

(21)

pg. 20 constituents (Deephouse & Carter, 2005). ErAs response to pressure, as rational beings, are likely to vary depending on the level of preparedness. Even then, it is not always clear if things are going to work according to plans. In other words, they have to embark on risky measures, deciding when to respond or stay silent (Maor et al., 2012). This is also the case when the media instils doubts about the potential for an agency to actually carry out their mandates. For example, a piece written in Politico EU highlighting the possibility of a potential struggle for the new financial EU regulators created in 2011 (EBA, ESMA, and EIOPA) can stand as a form of pressure (Wishart, 2011). The pressure stemming from the creation of ErAs therefore has an impact on reputation management.

Moreover, it is often believed that how one starts does not need to determine how they will finish, however what is often left out is the impact it would have nonetheless. In an environment with limited resources and relative reputation distribution (Deephouse & Carter, 2005), the ability to deal with pressure can determine what happens next. Consequently, unless ErAs are able to copy in the face of pressure, they face the risk of having to take a step back to recalibrate their reputational goals which has significant consequences. With survival as a key marker for institutions, one would expect that ErAs would be willing to do what needs to be done and adapt to successful models or consequently risk faltering.

Expectation 1: The more initial pressure an ERA faces, the more likely it would result in converging with the dominant reputation strategy.

Age:

Age with respect to ErAs refers to the time period in which they were created subsequent to the EU ‘agencification’ process. Deephouse & Carter state that the “length of time that an organization has been in business may be positively related to […] reputation” (2005, p.344). Consequently, the pressure of being relatively new can have an effect on the reputation communication and can therefore have an effect on the organizational behaviour. When ErAs are crafted, like starting a new school for the first day, one concern would be to start well. Starting well can be used to foreshadow what is to come in the future. Furthermore, dealing with a multitude of complex but interrelated constituents (Carpenter, 2010; Carpenter & Krause, 2012) can become a daunting task, because one cannot always please everybody at the same time. Hence, the burden that stems from making a good impression can take its toll on regulatory agencies if they are unable to see this initial reputation building stage through.

(22)

pg. 21 Nonetheless, if relatively new agencies are able to ride under the wings of pre-existing reputable agencies, the initial stages will see ErAs more focused on what matters which is delivering on their mandate and consequently have a fairly smooth sailing journey. At the same time, over-time, pre-existing reputable agencies can turn out to be a model that could come to be adapted by others creating similar structures. In view of this,

Expectation 2: Agencies classified as new are less likely to engage in reputational strategic change if they are not under pressure.

Expectation 3: Older agencies are less likely to engage in reputational strategy change.

Agency Independence:

In addition, it is often argued which form of control is effective to serve as an accountability mechanism against agencies (Busuioc 2009), but what is also of key importance is the level of formal independence (can also be referred to as de-jure autonomy) ascribed to each ErA upon inception. Agencies are linked with varying level of independence when it comes to their mandate (Groenleer, 2014), as a result, the power within an organizational field is unevenly distributed. Consequently, the shot at self-determination and possibly cultivating a reputation can be heavily influenced by formal independence level. There has been a plethora of discussions stemming from analysing the underlying impact of de-facto and de-jure autonomy (Groenleer, 2009), or put differently, the actual level of control that matters to regulatory agencies. Regardless, the initial stages of conception will be crucial to determining how the future of the organization will develop. With a limited level of control, agencies are able to hide in the shadows of political principals in the event that policy issues arise. In that sense, ErAs with lesser formal autonomy, while having their reputation, can tap into the reputation of its principals, thereby making it somewhat easier to function and shift blame (Hood, 2002, 2007, 2011; Maor & Sulitzeanu-Kenan, 2013; Moffitt, 2010). In that respect,

Expectation 4: The lower the formal independence, the less likely an agency would conform to the dominant reputation management during certain period in its development stage.

Expectation 5: The higher the formal independence, the more likely agency will choose to conform to the dominant reputation strategy.

(23)

pg. 22

Policy Saliency:

Furthermore, policy saliency is another key signifier for potential isomorphism in reputation management. Moreover, several studies have paid particular attention to the impact of policy/issue and agency saliency (Gilad et al., 2013; Maor et al., 2012; Wood, 2018) as a determining factor for agencies’ pursuit of reputation. The impact at the initial stages of agency life can prove equally vital as they would need to find a balance between the actual salience levels of a policy area with respect to the potential threat that can arise from having multiple constituents (Carpenter & Krause, 2012). The more the number of diverse onlookers, the riskier it becomes for agencies vis-à-vis their reputation goals (Carpenter, 2010). What is more, while all agencies may have been created on the backdrop of a salient policy area, the subsequent level of saliency after the creation of the ErAs can shed light on the pressure attached to varying regulatory agencies. Additionally, in the face of pressure arising from high saliency at agency’s infancy stages, because ErAs may not always get it right, an option would ultimately be to re-evaluate its bid for unique reputation strategy by adapting to a more successful strategy of communicating reputation. In this case, while risky, it increases the potential survival chances of agencies to consider changing reputational strategies to what appears dominant. Consequently,

Expectation 6: Agencies facing high threat from policy saliency at the initial stages of organizational life will be more likely to copy the dominant means of reputation management.

INTERNAL PRESSURE

Agencies can be internally fragmented (Black, 2001), but a positive external image can convey to their multi-layered audiences their potential competencies (Carpenter, 2010) and at the same time hide their flaws. Moreover, as iterated, regulatory agencies are not always aware of what to do especially when it comes to simultaneously pleasing the various multifaceted audiences (Carpenter, 2010). As a result, there is a need to branch out –discretely, if possible- and pull resources from external sources. In this case, it is drawn from agencies’ desire to deal with functional issues that may be unclear and could pose potential threat to its reputation, be it

(24)

pg. 23 perceived or actual. When agencies are not fully knowledgeable on what to do with regards to their task, it becomes imperative for them to adapt to other model in a bid not to look incompetent. Since reputation covers varying dimensions (Carpenter & Krause, 2012), another issue that may arise that may trigger agencies to consider getting help from outside is when there is a reputational threat, be it with regards to scientific output (Rimkute, 2018), or concerning the organizational culture (Bankin & Waterhouse, 2018; Yesilkagit, 2004). Organizational culture can matter for public agency and as a result needs to be carefully considered when building and maintaining reputation (Wæraas & Byrkjeflot, 2012). These issues can lead to an agency taking the initiative to solve the issue before it becomes a contentious subject for principals and affected others. A strategic response to the problems can stem from open cooperation with others (Busuioc, 2016) or being a part of multiple networks. By being part of networks, such as the EU Network initiative or the European Supervisory Authority (ESA), information are easily disseminated providing information for reputation seeking agents to copy, thereby helping them adapt to the environmental expectation. This does not mean issues like turf wars will not ensue (Busuioc, 2016).Nonetheless, the effect of internal pressure could potentially impact convergence or divergence on reputation seeking aspects. As a result of this,

Expectation 7: In the case of internal uncertainty, ErAs will be highly likely to be pulled towards the dominant reputational dimension.

Expectation 8: When ErAs are not under internal pressure, they are less likely to engage in reputation change.

EXTERNAL PRESSURE

Finally, when it comes to the influence of constituents mounting pressure over regulatory agencies, I believe this is perhaps one of the most effective that can yields result. External pressure can be likened to coercive pressure referenced by D&P, as institutions are pushed towards a certain organizational change (1983). This mostly concerns the ability of principals and important others to use resources and legal constraints as a leverage to impact an agency’s decision-making powers. In addition, what is understood is the fact that in dealing with external pressure, institutions are either able to engage passively or actively with important constituents (Oliver, 1991). It is thus expected, as with binary terms, that both response styles are not projected to manifest simultaneously and rightly so because it is either agencies do something

(25)

pg. 24 about a mounting pressure or not (Maor et al., 2012). Furthermore, what appears to be neglected is the fact that reputation also matters very much for principals and politicians (Busuioc & Lodge, 2017). No rational entity would like to be associated with an incompetent institution. Additionally, it is possible that influential constituents are highly likely to have a model in mind regarding what they expect from agencies and this can be based of existing institutions, What's more, ErAs, like other institutions alike have to react to pressure from groups such as its principals, the media, pressure groups and general public (Löfstedt & Way, 2016; Rimkute, 2018 . As mentioned earlier, this can vary from passive to active. In Oliver’s (1991) study, the varying strategic responses fit well with this consideration. The constraints as a bargaining chip can work to steer regulatory agencies into convergence, especially vis-à-vis the principal’s choice of action. External involvement, particularly from the principals would only occur on the ground that a risk assessment conveys that the impact of reputational loss could be detrimental. To predict external interference, the manipulation of and reference to the legal mandate and the resource allocation (Ashworth, Boyne & Delbridge, 2009) can offer compelling evidence for regulatory agency’s decision to copy others. Subsequently, depending on the overall impact of external pressure, it can potentially pave the way for homogeneity across reputation management strategy. In view of this, I suggest that,

Expectation 9: The higher the external pressure on ErAs, the more likely a change to their dominant reputational management strategy will occur.

(26)

pg. 25

4. METHODOLOGY

Having laid the theoretical underpinnings, this chapter will look closely at csQCA -which is the research design to be used in this study- what it entails, including the general benefits and the downsides attached with it. A brief summary on the important principles of the approach will then be provided. Following this, the outcome and condition variable will be conceptualized and operationalized, and in so doing, the research method is discussed. To recall, this study attempts to look closely at ErAs in a bid to uncover the pathways that has led to prioritising a dominant reputational dimension despite the push for uniqueness. First, a brief introduction on QCA is mandated.

4.1 Introducing QCA

QCA is a method which was initially coined by Charles Ragin in order to analyse and compare small to medium-N cases (1987). It serves as a suitable method to uncover necessary and sufficient conditions leading to an outcome (Ragin, 1987; Rihoux & De Meur, 2008). QCA has cemented itself as an innovative strategy piecing together quantitative and qualitative techniques. Therefore, the question of this study coupled with the number of cases available and the parsimonious advantage offers compelling reasons for the adoption of this method. Causality is not a straightforward phenomenon, rather QCA takes a different approach wherein a host of factors, referred to as set relations (Greckhamer, Misangyi, & Fiss, 2013), can play significant or minimal roles in explaining the presence or absence of an outcome of study. Besides, set relations for the most part provides more tangible answer to research questions. As Berg-Schlosser and colleagues puts, “QCA techniques strive to achieve some form of “short” (parsimonious) explanation of a certain phenomenon of interest, while still providing appropriate allowance for causal complexity” (2009, p.10). Furthermore, with QCA, the goal is to “generate some new insights, which may then be taken as a basis for a further theoretical development or for re-examination of existing theories” (Berg-Schlosser, Meur, Rihoux, & Ragin, 2009, p.16). In line with this understanding, the purpose of this study is to test and develop new theoretical links. Since similar trends exist with regards to reputation communication, it is important to uncover the set relations that brings about this outcome. Moreover the number of cases in this study resonates with that of (Busuioc & Rimkute, 2019) which looks closely at 20 EU agencies (see Appendix I). What remains is to introduce what

(27)

pg. 26 QCA entails, its advantages and disadvantages, and then conceptualize and operationalize the conditions and the outcome.

4.2 More about QCA

In brief, QCA which functions through the use of Boolean algebra (Marx & Dusa, 2011) is carried out through the use of computing programmes. This technique requires that a table with the combination of conditions for each cases be put forward with the values in binary terms. Binary terms simply demarcates between being a part of a relationship or not and is represented through the value of ‘1’ or ‘0’ (Marx, Cambré & Rihoux, 2013). Based on the combination particular to cases, a truth table is provided that conveys a configuration of brings about varying outcomes. Moreover, depending on the number of conditions, it would determine the complexity of the configurations. In addition, one of the benefits attached to QCA is based on the use of a truth table, which enables the creation of a formula or Boolean expression utilized to properly represent the cases under review and potentially the population. That said, generalizability can be an issue in QCA. As a result, Emmenegger and colleague have attempted to address this issue in their (2014) study. It is important to further note that with QCA, one of the benefits is in its ability to highlight necessary and/or sufficient conditions; INUS conditions can also be discovered (See Thomann & Maggetti, 2017; Toshkov, 2016). Finally, Boolean minimization sets the ground open for analysis, interpretations, and generalizations in light of the causal relationship discovered from the configuration.

4.3 The Incentives of QCA

The first advantage in using QCA is the fact that it can act as a bridging factor to inculcate qualitative (case-oriented) and quantitative (condition-oriented) analysis (Marx & Dusa, 2011, p.105). With this design, there is a possibility for different methods to contribute to the causal explanation and shed some light on the topic at hand (Thomann & Maggetti, 2017). This is crucial as there are studies that focus wholeheartedly on one aspect of analysis, thus limiting itself to certain advantages. This is not the case with regards to QCA, although there are potential downsides to using the specific forms of analysis vis-a-vis external validity (Thomann & Maggetti, 2017). Another challenge comes from measuring variables that are subjective in nature, though statistical studies also face similar issues.

(28)

pg. 27 Furthermore, the use of QCA serves as a commanding technique because of its ability to produce transparent, replicable and analytical results at each stage of a study (Berg-Schlosser et al., 2009). As a result of this, readers can follow with relative ease. In addition, because of the manner in which QCA technique conducts a study, there are possibilities to dissect and take into account different issues when considering analysing variables and the ensuing results. Therefore, it offers itself as a deeper and richer explanatory tool for scholars.

In a similar light, because of the possible different causal relationship that can be discovered -referred to as Equifinality (Marx & Dusa, 2011; Berg-Schlosser et al., 2009) - it becomes difficult to overlook cases without adequate prior knowledge. What is more, Equifinality makes it possible to locate the necessary and/or sufficient -as well as the INUS - conditions which can aid in highlighting possible causal mechanisms in play. This comes through the Boolean minimization technique where causal relationship is depicted to convey what is of paramount importance or not. It is also crucial to understand that the impact of different conditions can have asymmetric effects on the outcome (Berg-Schlosser et al., 2009), hence the need to dwell carefully in explaining the presence and absence of conditions.

Having said that, while QCA embraces the positive aspects of using qualitative and quantitative methods, it also takes with it some negative aspects. The context-based reliance of quantitative technique limits the external validity of the research, as can be the case with unit of analysis for this study. Nevertheless, this limitation can be fairly mitigated through close case studies (Emmenegger, et al., 2014). What is more, QCA is usually looked down on by scholars because QCA methodologists are usually thought to give unmeasurable values to concepts (Emmenegger, et al., 2014). Even with that, QCA offers itself as an ideal technique for investigating the suitability of the theoretical framework posited in this study; it enables the possibility of considering each conditions as a phenomenon on its own while also looking at them in relation to other conditions (Emmenegger, et al., 2014; Berg-Schlosser et al., 2009).

Table 1 Key Operators for Boolean algebra

SYMBOLS MEANING ADDITIONAL EXPLANATION

“+” Logical ‘OR’

“*” Logical ‘AND’

“0” CAPS Presence of condition/ outcome

“1” Caps Absence of condition/ outcome

(29)

pg. 28 (For a succinct instruction on the use of csQCA, look at Rihoux & De Meur, 2009)

4.4 Conceptualization and Operationalization

In this section, the aim is to transparently approach the ways in which the measurement of the operationalized conditions will be carried out and where data will be retrieved from. For a summary of the operationalized concepts, how they have been measured and where these data have been retrieved, see ‘Appendix IV’.

4.4.1 Outcome

ErAs are divided when it comes to their respective tasks which as a result is expected to play a role in agency’s reputation management strategy (Carpenter & Krause, 2012). Busuioc and Rimkute in their study highlighted the importance of direct (enforcement/supervision and decision-making) and indirect regulation powers (information and advice dissemination amongst others) in defining regulatory agencies (2019). Taking it a step further, because of their main regulation power, it is logical to expect that ErAs will focus primarily on communicating this aspect of their task in their respective reputation management. For example, EU-OSHA which is expected to provide information (Vos, 2018) and ensuring worker safety and health will ideally be interested in mainly communicating their procedural or moral reputation management. In fact, this appears to be the case when considering what EU-OSHA communicated to its constituents upon its inception in 1996 (Busuioc & Rimkute, 2019). Over time however, with the exception of few ErAs, there has been a focus on performance and technical reputation making them the dominant choices (See figure 2). Perhaps reputation strategy suffers from the lack of an encompassing concept, nonetheless, this is the next best thing until further encompassing concepts are theoretically and empirically observed.

When it comes to the two dominant reputation dimension for ErAs, these operationalized concepts have been directly extracted from Busuioc and Rimkute’s (2019) study. The data was achieved by scrutinizing the selected ErA’s documents –e.g. annual reports, multiannual plans, work programmes, etc. - from the time of inception (based on availability) and parsing through for keywords that represented these dominant dimensions. Data was computed through the use

(30)

pg. 29 of Wordstat which aids in carrying out a qualitative content analysis (Busuioc & Rimkute, 2019). While the data was analysed separately –and while all ErAs are not included for practical reasons which can significantly impact the findings of this research, for the purpose of this study it would be important to consider the reputation management for each ErA in relation to each other. ErAs function under the same environmental field therefore their actions can have a direct and relative effect on each other.

Furthermore, because the interest is to look at the variation in the choice to communicate a dominant reputation strategy, the year 2015 provides the optimal rise of performative reputation communication surpassing technical reputation (Figure 2). Therefore, this serves as a good point of reference to measure potential isomorphism retrospectively. In view of this, for the purpose of this research, binary terms will be utilized to convey if any reputation strategy change was experienced by ErAs between their respective time of establishment and in the year 2015. A change to the dominant reputation is represented in binary term as [1] while the absence of change or the constant communication of a distinct reputation strategy represents [0].

4.4.2 Condition Variables

In line with the previous chapter, initial pressure which is divided into formal independence, policy salience, and age, together with internal pressure and external pressure are of key importance to isomorphic tendency. This concepts will be discussed and operationalized with a view on how they will be measured (See Appendix II & III).

Initial Pressure

The pressure at the establishment phase of ErAs as iterated can be divided into 3 distinct yet interrelated areas. The typology utilized here can be measured separately to uncover the varying impact on the outcome of expectation. Moreover, it is important to bear in mind that this form of pressure dwells solely on the establishment period, as such, there is a potential for variation to occur when compared to the latter periods. For example, saliency level could either increase or decrease over time and formal independence may interact more frequently with de-facto autonomy during the ongoing control period (Busuioc, 2009). That said, what is to follow is how these factors have been operationalized and accordingly measured.

(31)

pg. 30

Agency Formal Independence:

Upon the creation of ErAs, as mentioned earlier, they are inscribed each with different levels of formal autonomy (Wonka & Rittberger, 2010). This is crucial psychologically and physically because it can determine what each ErA can or cannot do (Groenleer, 2009; Wonka & Rittberger, 2010). This is not to dispute the impact of de facto autonomy (Gilardi & Maggetti, 2010); it is possible that de facto autonomy evolves over time (Groenleer, 2014) so the impact is not immediately felt. Furthermore, an understanding of Groenleer (2009) conveys the importance of formal independence with regards to future goals which cannot be overemphasized. With decentralized agencies, it is somewhat difficult to adequately quantify and categorize the level of formal autonomy because the principals, upon creation of the ErAs, have attributed ambiguous mandates (Groenleer, 2009). This is potentially to enable broad acceptance by the member states, moreover, over time it could become clearer. Irrespective, to be able to demarcate between the level of formal autonomy, placing agencies as fitting with high formal independence [1] and low formal independence [0] will provide a necessary base for cross analysis.

While the formal independence score to be utilized is extracted from Wonka & Rittberger 2010’s study, there were few agencies not included in their measurement which required that I calculate the formal autonomy scores for the omitted agencies. Again, through careful consideration of measurement scales used by Wonka and Rittberger (2010), I used the calculated weighted score from two out of four criteria as the other 2 criteria were not quantitatively significant owing to its numerous subsections. Moreover, the omitted criteria also calculated as part of its subsection some key attributes that would have been significant for and reflective of other conditions such as external pressure or policy saliency, thus limiting redundancy was vital.

In addition, because autonomy can be a fairly relative concept with regards to agencies, an average independence score was utilized to demarcate agencies with either high or low formal autonomy (see Appendix II).