Developing Agencies and Their Organizational Reputation: A research of reputational behaviour of the supervisory agencies EBA, EIOPA and ESMA

(1)

Public Administration E&G 2018/2019

Developing Agencies and

Their Organizational

Reputation

A RESEARCH OF REPUTATIONAL BEHAVIOUR OF THE

SUPERVISORY AGENCIES EBA, EIOPA AND ESMA

MASTER THESIS

PUBLIC ADMINISTRATION

TRACK ECONOMICS & GOVERNANCE STUDENT NAME: JACOB ZWAAN STUDENT NO.: 1550934

INSTRUCTOR: DR. RIMKUTĖ, D. SECOND READER: DR. BUSUIOC, E.M. FINAL VERSION

(2)

1

Abstract

The ongoing agentification of the European Union is affecting - or even shaping - the European regulatory state. In the wake of the financial crisis, this agentification has taken shape in the formation of the European System of Financial Stability, an architecture that houses three financial European Supervisory Authorities. Within the framework of organizational reputation, EU agencies seek to establish and manage their reputations with a multiplicity of audiences. By doing so, they seek to strengthen their roles in various institutions. By balancing four dimensions of organizational reputation, EU agencies seek to strategically position themselves towards these audiences and institutions, by carefully employing strategies and communications that emphasize these dimensions and focus on their various specific audiences. So far, scholarly research on the perspective of organizational reputation is limited. With little literature available and the limited availability of thorough analysis of reputational behaviour of specific policy fields, we will seek to fill the research gap by applying this perspective on the financial ESAs. This research will assess the theoretical background of reputational behaviour applied to the reputational development of financial agencies in the ESFS. Analysis is done via a qualitative and quantitative approach of how the agencies have developed their reputational behaviour in their short existence. The development of reputational dimension is put in relation to external threat (i.e. Brexit), in order to map out the effects of these threats on reputational behaviour.

(3)

2

1. Introduction

The foundation of the European System of Financial Stability (ESFS) on January 1, 2011, in the wake of the financial crisis, is one of many examples in the ongoing agencification of the European Union. This architecture encapsulates the European Systemic Risk Board (ESRB) and the three European Supervisory Authorities (ESAs), namely: the European Banking Authority (EBA), the European Insurance and Occupational Pensions Authority (EIOPA) and the European Securities and Markets Authority (ESMA). A strong range of powers and capabilities are granted to these agencies, allowing them to draft binding decisions for national competent authorities within their field. Their scope ranges from (quasi-)rulemaking to supervisory authority.

The general purpose of the ESFS is ensuring control and supervision of the EU’s financial systems. It does so through these three authorities, cooperating through a joint-committee, and the ESRB as general risk board, which is hosted in the broader organization of the European Central Bank (ECB). These ESAs are enforcing networking organizations with quasi-rulemaking responsibilities, which operate under the common structure of the ESFS. There is a comparable governance structure for all the agencies within this ESFS framework, allowing for a good comparison of the agencies. Within this research, we are able to analyse the full span of the agencies’ relatively short existence from foundation to present. In this span, we are able to uncover trends, notable changes and differences for the seemingly comparable agencies. This research aims to analyse this ESFS in the framework of reputational behaviour. These agencies have a comparable governance structure through their common foundation, but differences in specialized tasks and threats within this framework allow for comparative research. Maor, Gilad & Bloom (2013) indicate a relation of strategic communication in reputation management, as a large portion of threats towards organizational reputation are based on external audiences. This strategic communication behaviour in relation to their external audiences of agencies will play a central role within this research.

Busuioc & Rimkutė (2019) indicate a more ‘natural’ shift within the reputational dimensions over a larger timeframe for European Agencies in general. Agencies have become more aware of their ‘reputational assets’ and adapt their behaviour towards this. Agencies may cultivate their reputation by choice, this choice often depending on their audiences and relevant risks. In this research, we will analyse how this behaviour has evolved in the specified cases of the financial agencies. The central questions within this research will address the differences

(5)

4 between the agencies in their reputational behaviour choices, within the perspective of four dimensions of reputation. These differences in their behaviour choices are possibly driven by their various differing specializations, utilization of their governance structures, external relationships, and external threats that influence these agencies.

The theoretical framework of reputational behaviour (Carpenter, 2010; Carpenter & Krause, 2012) will provide the foundation for our analysis of the four dimensions of reputation, namely: technical reputation, performative reputation, moral reputation and legal-procedural reputation. For an organization, reputation is a strategic resource, which they need to develop and maintain towards their audiences in order to validate or enhance their operations. The development of these reputational dimensions within the agencies, and especially the differences between the ESAs, will be analysed. The main question of this research is: In which way have the ESAs

within the ESFS framework developed their reputation over time and what explains their differences in reputational behaviour?

The framework of the ESFS and close cooperation of the networking agencies housed within this framework provides a chance for comparative research. Possible differences in reputational behaviour and the drivers behind that may expand on existing literature. We will do so by providing insight in these different drivers of reputational behaviour within seemingly similar organizations. Even slight differences between the ESAs in the evolution of their reputational behaviour may have strong underlying explanatory values that add to the background of the reputational approach. This research aims to uncover the communication and presentation the ESAs utilize towards a broad public to build and maintain their reputation. By analysing annual reports and work programmes, the evolution of the reputational behaviour of each of the agencies is mapped out in the analysis. In broader perspective, the research of the ESFS agencies provides an insight in the manner in which they report towards and inform the public. It also uncovers the way in which they utilize their (reputational) assets within the scope of their different audiences and what their focus is within this perspective. This research aims to build further upon the scarce literature of the way European agencies strategically position themselves within their communication and the connected perspective of reputational behaviour in EU agencies. Continuing upon agency perspectives of Majone (1998), Carpenter (2010), Busuioc & Rimkutė (2019), this research aims on filling the research gap on the mapping of reputational behaviour of specific EU agencies.

The structure of this research is as follows: Firstly, the theoretical framework of organizational reputation, reputational uniqueness and reputational threats will be presented. This framework

(6)

5 will form the basis of the research question posed and be the foundation of the analysis. Secondly, we will draw a methodology and a short presentation of the individual cases. The methodology will also provide a description in the substance and manner of the documental research that forms the quantitative part of this research. Fourthly, the results for the analysis of the agencies will be presented. Fifthly, a discussion regarding the roles, the strengths and the relationship of the separate authorities will find place on basis of these results through the scope of our framework. Central questions will be: how have these organizations developed their reputations over time, where are possible breaks in expectations and how can these be explained? Finally, we shall conclude our research with a summary of our main findings and an answer to our research question in the final chapter.

(7)

6

2. Theoretical Framework

2.1. Organizational Reputation

The framework of organizational reputation as described by Carpenter (2010) and Carpenter & Krause (2012) is treated in relation to organizations and their audiences. An organization operates on different dimensions of this reputational spectrum, being technical, performative, moral and procedural reputation. These reputations may correspond with different audiences (Carpenter & Krause, 2012:27) and an organization may choose on which specific audience and therefore reputation they desire to focus. An audience is any party, individual or collective instance, that observes and judges an agency and its behaviours (Carpenter, 2010:33). In a way, organizational reputation defines the relationship between an agency and a specific set of audiences (Carpenter, 2010:45). Reputation does so, because agencies are able ‘choose’ whether they focus on one or more of the given dimensions of reputation – the preference of one or more of the four forms of reputation may complement the form and objectives of an organization, both in their internal and external activities.

It is important to note that organizational reputation knows overlap between the four different dimensions (Carpenter, 2010:47) and that it is possible for certain reputations to compete internally, or for some agencies to compete over the same reputations. Reputation is an asset; which organizations often seek to expand or maintain. Multiple audiences often support the decision to focus on multiple dimensions of reputation (Carpenter, 2010:7), as different audiences uphold different expectations and standards of what ‘good’ reputation is. Reputation as an asset promotes an organizations’ regulatory powers with these audiences, as their regulatory outputs have a stronger support (Carpenter & Krause, 2012:26-27; Maor, Gilad & Bloom, 2012:601). According to Maor et al. (2012:601-602), not only do organizations with a good reputation have a stronger support in prospect, but also in retrospect. The output of these organizations is perceived as more just and sound and therefore less susceptible to external scrutiny. Reputational management thus strengthens organizations’ actions in the input and output phase. The four dimensions of reputation and their underlying expectations and particular details are listed below.

2.1.1. Technical Reputation

If an organization decides to focus on their technical reputation, they often house expert knowledge and seek to be the qualified entity in their field of expertise (Carpenter, 2010:46). The organizations with ‘good’ technical reputation hold an identity of technical and methodological prowess within their policies and regulation. This is particularly relevant for

(8)

7 autonomous networking agencies in the EU. An organization that seeks to cultivate a good reputation on the technical aspect, often seeks to become an authority within its field. It attempts to block out competition, by ‘proving’ their evidence is most structurally and scientifically sound (Maor, 2007: 963, 974). Creating this ‘gold standard’ creates an advantage for an agency to exert both internal and external pressure within and through networks. The choice for a framework within broader political context shapes a status quo and work in the advantage of the agency. High technical reputation embeds an organization within their fields of operation. Technical reputation is backed by a larger extent of scholarly validity and a structural soundness in scientific reporting towards audiences. Technical Reputation has the advantage of being able to speak to towards a multiplicity of audiences and is therefore utilized for many organizations (Carpenter & Krause, 2012:27-28). This does also create a complexity within the balancing of these audiences, as satisfying one audience may upset the other.

Technical reputation and the importance thereof for rulemaking and risk assessment is a given for such regulatory agencies. Technical prowess is of much importance within this field of regulation, as much is policy is based on data, assessment thereof and the bundling of scientific knowledge in order to produce good regulatory output (Rimkutė, 2018:70-71).

2.1.2. Performative Reputation

Performative reputation is grounded in the judgement of the performance of an organization by its audiences. An organization is judged in the manner it responds to internal and external shocks, as well as the extent as to which it actively and successfully achieves its goals and tasks (Carpenter, 2010:46) set by external audiences. Audiences ‘test’ organizations on their performance in asserting their power within regulatory frameworks – audiences ask themselves the question if an organization is competent enough to take up a task. Performative reputation is in essence the way in which audiences perceive an organization capable of standing up to their assigned tasks, but also their resilience towards external inputs.

The performative dimension is highly based on co-operation and coercion that exists between these co-operations (Carpenter 2010:47, Carpenter & Krause 2012:27-28). Competences and the strength in which they manage to communicate these competences between layers is essential within this reputational dimension. The capacity to coerce other authorities to comply is a key factor for certain organizations.

(9)

8

2.1.3. Legal-Procedural Reputation

Legal-procedural, or simply procedural, reputation focuses on the regulatory processes and procedural soundness. While there is some overlap with the other reputations, procedural reputation looks at the justification of the process. If an organization takes regulatory action, are these actions thorough and justified? Have they followed due processes, and do they adhere to the set rules? This form of reputation mainly speaks to courts, but also scientific audiences (Carpenter, 2010:47). A thorough process may add to both moral and technical reputation. Due to this overlap, it is important for organizations to follow these procedures. Apart from its scientific output or performance, a ‘just’ or fair organization is seen as a more legitimate entity.

2.1.4. Moral Reputation

Moral reputation refers to the organizations’ response to their clientele. Is what an agency does and how they do it ethically acceptable (Carpenter 2010, 46)? This form of reputation has great overlap with other reputations. Is an organization’s scientific prowess ethically sound? Is their way of organizing transparent enough? Moral reputation is a factor that often comes back in broader discussions. Maor, Gilad & Bloom (2012), for example, research communication within the Israeli Banking regulation. Moral reputation is largely about external communication towards audiences and the way this is done. In order to maintain their reputation, actors choose to remain silent. If an organization seeks to expend their reputation, they are more likely to break silence and promote their ways of working (Maor et al, 2012:601). This goes up for multiple forms of reputation, but the maintenance of moral reputation often goes hand in hand with solid codes of conduct – organizations are inclined to remain ‘good’ in their behaviour.

2.2. Reputational uniqueness and threats

In the extent of organizational reputation lies the concept of reputational uniqueness. This reputational uniqueness is described as an organization’s capability to produce outputs that are not found anywhere else (Carpenter, 2001:5; Maor & Sulitzeanu-Kenan, 2015:33-34).

European agencies adapt to this uniqueness and utilize this knowledge in relation to their reputational concerns (Maor & Sulitzeanu-Kenan, 2015:33-35). Reputational uniqueness steers the direction of agencies’ reputational behaviour, as agencies may tailor their output in regulations and strategic behaviour (Maor, 2014:25) and communications (Maor et al., 2013:601) towards their agencies’ uniqueness. Maor et al. (2013:602) underline the central role of communication strategy within reputational behaviour. Their communicative strategies may also be connected to agencies’ other output. How does their communication relate to their response to external inputs (Maor et al, 2013:602)? While their research is aimed on national

(10)

9 agencies, the dynamics of these communication strategies and relation to material action in response to external signals can be used in research on EU agencies.

The quasi-rulemaking ESAs share their common grounds but are also likely to diverge within their behaviour. This is due to their specializations and differing external environments. Each of these agencies has their reputational uniqueness and, while there is overlap between the agencies’ fields, has different audiences. Each of these agencies has to adapt to specific challenges, resulting in the ESAs having different reputational threats and risks to cope with. Organizational threats (and risks) are defined by Gilad, Maor & Bloom (2013:452) as the audiences’ opinions that threaten agencies’ established reputation and relevant outputs. Reputation management (Gilad et al, 2013:453-454) provides the answer to these threats. An agency may strategically position themselves in communication and other outputs in order to defend themselves from relevant risks and threats.

Maor (2014) and Maor & Sulitzeanu-Kenan (2015), Gilad et al. (2013) have analysed the manner in which agencies respond when faced with reputational threats. Agencies may seek to change their performances through their own evaluation thereof, as a result of exogenous threats. This ‘responsive change’ (Maor & Sulitzeanu-Kenan, 2015:41) is more volatile when reputational threat towards an agency is high. An agency responds to reputational threat in different manners, it changes its output accordingly and allocates its resources in different manners when threatened. This occurrence may shift the reputational focus of an agency, in order to defend itself from the existing reputational threats.

2.3. Reputation within the ESAs

Carpenter (2010:46-47) underlines the importance of enforcement of rulemaking organizations towards national competent authorities. Through the non-horizontal framework in which the ESAs operate as a quasi-rulemaking agency through the Commission (Busuioc, 2013:113) this becomes even more important. EU agencies must strategically position themselves with their actions and within their communication in general, but specifically towards national counterparts, in order for the agencies’ output to gain support with their audiences.

In relation to the arguments of reputational uniqueness, the ESAs should be seen as (quasi-)rulemaking economic, advisory and enforcing agencies – this given role inclines the agencies to operate in a given way regarding reputational behaviour. Strategic communication is of major importance to these agencies. Due to their regulatory roles, these agencies are more inclined to emphasize the reputational behaviour than their counterparts (Rimkutė, 2019:22).

(11)

10 As advisory agency the technical focus can be seen as a baseline of a European agency reputational focus in general (Busuioc & Rimkutė, 2019:29). This argument is discussed in an early stage of the European agentification by Majone (1998:16;22), the delegation of policy-making powers towards neutral experts via an agency ‘naturally’ leads to an entity that primarily upholds high technical standards. This tool for an effective and efficient process is translated in the case of the ESAs.

Hypothesis 1: Technical reputation is most important for the ESAs. Their financial regulatory nature and their depoliticized nature incline them to focus more on the technical dimension over the other dimensions for the agencies.

Next to the importance of technical reputation, the agencies secondly have their focus on the performative reputation (Busuioc & Rimkutė, 2019:30). The supervisory and enforcement roles that the ESAs hold are an important reason for them to focus on this dimension. Strong management and propagation of this performative dimension allows the agencies to uphold a stronger role towards national competent agencies (Carpenter, 2010:46). For the ESAs it is necessary to ‘show their teeth’ towards their national counterparts in order to legitimize themselves towards the national competent authorities. Performative reputation is of major importance in this enforcement role – good performance and indications of fair play create leverage over these national competent agencies.

Hypothesis 2: The ESAs focus on performative reputation second to technical reputation. Due to the nature of their supervisory/enforcement tasks in relation to national competent authorities, a strong focus on the performative dimension is necessary for the agencies.

The current ESAs structure gives the agencies a great set of power, but these are still embedded within the framework of quasi-rulemaking (Moloney, 2011; Busuioc, 2013) through the Commission. The Meroni case (European Court, 1958) effectively provides these standards of limitations towards the powers the Commission can delegate upon agencies in the forms of executive and discretionary powers. The powers that are present within the ESAs are still strong, as the Commission only has restricted room to amend or dismiss regulations brought up by the ESAs (Busuioc, 2013:117).

Somewhat circumventing the Meroni standards, are the guidelines and recommendation provided by the ESAs, allowed by their legal structures. These are applied towards national competent authorities and are communicated within the annual reports and thus communicated towards a broad public (Busuioc, 2013:118). The guidelines and recommendations are not

(12)

11 legally binding and are therefore not encapsulated directly within these standards. These guidelines and recommendation fall in a grey area and allow the ESAs to have a stronger presence within their frameworks, towards the Commission and other involved national authorities. In line with the arguments of reputational threats, the threat of the ESAs technical audiences is likely to decrease over time within this developing framework in which they have ‘secured’ their technical roles and are more likely to shift towards the performative dimension in order to emphasize their supervisory tasks.

In addition to this, (Busuioc & Rimkutė, 2019:29) argue that the trend of New Public Management is promoting the performative dimension in EU agencies. The focus of the NPM trend in EU agencies has shifted the focus towards efficiency and thus performance measurement and reporting within their communications. The expectation is that over time, the performative dimension will overtake the technical dimension within the ESAs as they increasingly focus key performance indicators and on their enforcement roles.

Additionally, Busuioc & Rimkutė (2019:30) finds evidence that young enforcement agencies are likely to have an increased focus on the procedural dimension initially, but that this decreases with the age of these agencies. Young enforcing agencies may choose to do so in order to settle their roles in due process towards both national competent authorities and the broader public. Relying on good procedures in enforcement actions ensures this agency to settle within its network, once this foundation is laid out this dimension requires less attention (Busuioc & Rimkutė (2019:24).

Hypothesis 3: Over time, performative reputation will become more prevalent than technical reputation. The nature of the agencies will shift focus to its national competent authorities, as the ESAs are inclined to secure their roles relative to national competent authorities.

Hypothesis 4: The ESAs focus on the procedural dimension structurally declines over time, due to their aging as enforcement agency. The procedural dimension is of higher importance at the foundation of years an agency, but becomes of less importance when an agency settles its role towards national competent authorities.

Rimkutė (2018:12-13) argues that the extent in which an agency is faced with reputational threats has direct impact on response strategies. The agencies are more drawn towards communications and showing their ‘just’ ways of operations. According to Maor et al. (2012:602) agencies are faced with a binary choice of communicating or remaining silent. However, if communicating, the content of these communications does matter. Due to this,

(13)

12 communication is not necessarily a binary choice – as for the case of the European agencies public communications are constant, yet adaptive. This adaptive communication translates into different focus on dimension within this communication.

Our expectation is in line with Maor & Sulitzeanu-Kenan (2015) and Gilad et al. (2013), which state that agencies are prone to ‘responsive change’, yet the content and context of this response is of major importance. The amount and magnitude of reputational threats lead to a responsive change, which creates a difference in the allocation of strategic resources within agency communication and other outputs. For the ESAs this leads to the expectation that the EBA, in the wake of Brexit, alters its behaviour more heavily than its EFSF counterparts to this exogenous threat. Following Rimkutė (2018:23) we expect an increased focus on both the technical and performative dimension by EBA relative to the ESMA and EIOPA, in order to strengthen their legitimacy with their broader audiences. Rimkutė (2018:21) argues that the output-oriented EU regulatory state inclines agencies to rely on their base competences and expertise, thus on the technical and performative dimensions. The agencies’ choice for the technical and procedural dimension rests on the assumption that, as a defence to threats, emphasis will be on the technical and operative values of the EBA (Maor, 2007:975). The scientific ‘gold standard’ and an agency’s baseline technical competence, as well as performance measures are likely emphasized in uncertain policy situations. Grasping back to technical prowess and key performance is a relatively easy way to legitimize an agencies’ position.

Hypothesis 5: The higher institutional threat on EBA due to Brexit as external shock, inclines the EBA to focus on different dimensions relative to the ESMA and EIOPA. EBA is expected to strengthen their legitimacy through output orientation, resulting in an increase of the technical and performative dimension.

(14)

13

3. Research Design

3.1. Cases

3.1.1. General

The ESFS is the vessel for regulations and standards in the form of a substantial framework that was proposed by the European Commission, Parliament and Council. The ESAs within this framework have a comparable structure, each focussed on a specialized market within the financial field. The ESAs are largely networking structures which employ the knowledge and structures of their national counterparts in the formation and the application of policies within the framework of EU member states – they are able to formulate (mandatory) guidelines and rules for these national competent agencies.

Due to their nature of enforcement (Carpenter, 2010:47) and structure of their quasi rulemaking (Busuioc, 2013:113-114) through the Commission, the ESAs are inclined towards a more technical reputation, followed by performative reputation. Within communication they have to position and prove themselves towards national competent agencies, which are a technical audience that has to adhere to the chosen procedures by the ESAs.

The ESAs are established within the Joint Committee (2016). While the three authorities that fall under its scope are each individually accountable to the Parliament and Council, the Joint Committee is the body through which the authorities meet their obligations when it comes to issues that exceed their regular field of operation, such as cross-sectoral issues. The individual operations are anchored in the ESAs regulations: Regulation (EU) 1093/2010, Regulation (EU) 1094/2010 and Regulation (EU) 1095/2010.

3.1.2. European Banking Authority

The EBA finds its foundation within Regulation (EU) 1093/2010 and was amended via Regulation (EU) 1022/2013. These founding Regulations ensure the EBAs operation according to established Directives and agreements on the operation of banks within the EU. Article 16 of the founding regulation is a compliance model for the national competent authorities, which models towards the concept of ‘comply or explain’.

EBA is the only agency of the ESFS located in London currently. In late 2017 EBA has received the additional task in providing guidance to authorities and institutions that are to relocate with regard to Brexit. In response, the EBA is tasked with the analysis of policy and risks in relation to Brexit for authorities. Additionally, EBA monitors subsequent financial risks of Brexit for

(15)

14 the EU as a whole (EBA, 2017). To do so, EBA relies on its network of national competent authorities.

As for the governance structure of the EBA, it consists of the Board of Supervisors (BoS), the Management Board (MB), and delegation towards the other ESAs. The EBA also has a delegation towards the ESAs Joint Committee. The BoS is the main decision-making body of policy decisions within the EBA. Decisions related to the resolution of financial institutions are delegated from the BoS towards the Resolution Committee, which observes and reviews draft regulations and guidelines. The BoS consists of the EBA Chairperson and heads of member state banking supervisors. The ResCo consists of a Chairperson and heads of national resolution authorities.

3.1.3. European Insurance and Occupational Pensions Authority

The EIOPA finds its foundation within Regulation (EU) 1094/2010. EIOPA is tasked with the protection of consumers, the protection and harmonisation of consistent and effective supervision over the scope of the full EU and a strengthening of coordinated supervision. EIOPA holds a similar structure as the EBA in supervising and management, but in addition to this EIOPA has two supplementary Stakeholder Groups and a review panel. The Stakeholder Groups consists of academics, industry and consumers representatives for both insurances and occupational pensions and are an additional democratic and controlling mechanism built in the governance structure op EIOPA. The Review Panel is an additional panel consisting of representatives of national competent authorities and acts as a peer review on supervisory practices, ensuring fair implementation and monitoring of policy.

3.1.4. European Securities and Markets Authority

ESMA finds its foundation in Regulation (EU) 1095/2010, it is tasked with the protection of the investor market and the harmonization of the rules on these financial markets. Its governance employs a similar structure as the other ESAs with some minor differences. ESMA employs the same governance structure as the EBA and EIOPA. It also has a representative stakeholder’s group in the same function as EIOPA’s. This group is consulted for reflection on draft standards and guidelines proposed by ESMA.

In addition to this, ESMA houses the Standing Committee of national representatives supported by the ESMA staff. A strong network is prevalent within ESMA. Within the Standing

(16)

15 Committee draft rules and guidelines of ESMA are made and reviewed in cooperation with national counterparts.

3.2. Methodology

Reputation is a concept that is strongly related to the multiplicity of audiences of an agency. Due to this nature of reputation, we have chosen to research annual reports and work programmes of the chosen agencies. These productions are mandatory and comparable throughout the different ESAs and are written to communicate on a large scale, speaking to this multiplicity of audiences of an agency. The reports seek to communicate to a broad set of audiences, as this information is publicly available (Busuioc & Rimkutė, 2019). These annual reports also show a unitary output of an agency, as it is a product of internally conducted discourse.

The ESAs are advisory and enforcing agencies which house themselves in an institutional network of politics and national competent counterparts. We will address specific trends for the ESAs on average and further break down this trend per agency, to see where differences may occur and finally what the underlying reason for this discrepancy in reputational behaviour is. Busuioc & Rimkutė (2019) have analysed a broader set of agencies and their reputational behaviour in relation to the logic of the EU regulatory state. They have done so by conducting an analysis of the annual reports of these agencies across history. Their analysis is a quantitative analysis, in which keywords are deductively derived from the reputational dimensions of Carpenter (2010). For this analysis, we will maintain the same keywords and analyse the data over time (Table 1). The keywords presented in the table are non-exhaustive but capture a great portion of the four separate dimensions by utilizing specifically characterizing terms for these dimensions. The build-up is as follows:

Technical reputation captures an agency’s ability to provide strongly reasoned outcomes. Rigorous research processes and structurally sound productions (Carpenter, 2010; Maor, 2007). The keywords provided indicate a focus on these scientific and structural processes.

Performative reputation is mainly aimed at the ‘business’ side of agencies. It is strongly oriented towards active output and a strong proactive role within their frameworks. Audiences and agencies that are oriented towards performative reputation are aimed towards strong (regulatory) action (Carpenter, 2010). The keywords provided indicate a focus on enforcement, result-orientation and decisive action.

(17)

16 Legal-procedural reputation is a process-oriented dimension. It holds tight towards regulatory structures and standards that are implied within the foundation of the agencies (Carpenter, 2010). These keywords are oriented towards these thorough processes and adherence to rules and procedures. Legal-procedural reputation is rooted in most agencies and shows overlap with the other dimensions. The keywords provided indicate a focus on the legal process and orientation towards internal processes.

Moral reputation is aimed at ethical and moral justification of agencies and their output. Moral reputation is often rooted within external communication and based on the trustworthiness of organizations (Carpenter, 2010; Maor, Gilad & Bloom, 2012). The keywords provided indicate a focus on this external environment and an orientation towards integrity and honesty, moral and ethical values.

Keyword Categories Keywords

Technical reputation reliab*, test*, analy*, assess*, calcul*, data, evidence*, examin*, expert*, investigat*, knowledge, likelihood*, methodolog*, model*, profession*, qualitat*, quantify*, quantitat*, research, rigo*r*, robust*, science*, scientif*, studi*, stud*, technic*.

Performative reputation deliver*, enforce*, result*, achieved, achievement*, action*, adopt*_decision*, application*, assertive*, complianc*, comply, effecti*, efficien*, goal*, improv*, inspect*, KPI* [Key Performance Indicator], objectives, oblige*, outcome*, output*, performance*, restrict*, success*, timely*, target*.

Legal-procedural reputation consult*, formal*, independen*, protocol*, process*, rule*, access_to_document*, access_to_information, appeal*, conflict*_of_interest*, control_standard*,

control_system*, declaration*_of_interest*,

internal_control*, internal_operation*, internal_system*, judicial*, legal*, liability, management_standard*, management_system*, procedur*, provisions, requirement*.

Moral reputation Protect*, values, committed_to, common_interest*, consumer*, credibility, dialogue*, engagement*, ethic*, flexibl*, good_governance, honest*, inclusiv*, integrity*, moral, openness*, precaution*, public_interest*,

respect_for, safeguard*, societal*, transpar*, trust*, users.

Table 1. The operationalisation of reputational dimensions (Busuioc & Rimkutė, 2019)

Due to the relatively short time since the establishment, 9 years, of the ESFS agencies the number of documents is low. At the time of research, annual reports are available from 2011 through 2017. In order to expand our scope, the research will also analyse existing work programmes. For each agency, these work programmes are now available from 2012 through 2019. These are also publicly available documents, thus corresponding with a big audience. The

(18)

17 annual reports and work programmes are deemed as comparable documents. These work programmes are however more abstract. These are a prospective product of what an agency seeks to achieve and how they aim to do so, as opposed to the retrospective nature of the annual reports. In this research, we have chosen to work with the average results between the annual reports and the work programmes. The combined output of these documents shows the agencies’ retrospective and prospective goals and ambitions. The weighted average of both these documents create a comparative tool which encapsulates most of the agencies’ broader public output.

For the agencies we have collected 7 annual reports and 8 work programmes per agency. However, for each agency we will drop the first work programme, so that we may compare these in line with the annual reports. The year of publication on the annual reports and work programmes would otherwise have a mismatch. Each of the reports and programmes will be provided with a year of publication as a variable. When working with the year of publication, we can see how the agencies position themselves in their reputational behaviour in a specific year. This leaves us with the work programmes 2013-2019, respectively published in 2012 through 2018. These years are in line with the publications of the annual reports 2011-2017, respectively published in 2012 through 2018.

For our last hypothesis we analyse the potential break that occurs when Brexit preparations start. The referendum took place on June 2016 and its aftermath is likely to influence the ESAs. Specifically, the EBA is expected to respond heavily to Brexit, as they are based in London while Brexit occurs. The Brexit aftermath is expected to show its influence on the data from late 2016 and onwards, effectively translating its effects into the annual reports of 2016 and 2017 and the work programmes for 2018 and 2019, thus translating its effects to publications years of 2017 and onwards. For the measurement of this hypothesis, we will analyse the differences that occur in publication year 2017 and further. At the time of writing the negotiations between the EU and the UK on Brexit are still unsure. Both the deal and no-deal scenarios are still open. The effects of Brexit on reputational behaviour are therefore still prevalent on 2018 publications. The young age of the agencies limits the available datapoints and the number of public documents in the form of annual reports and work programmes to be used for the dataset. In each of the cases, we will report a table with the development of the use of reputational dimensions based on these documents.

For the analysis we have researched the documents, using the dictionaries in Wordstat according to Table 1, thus using the provided keywords (Busuioc & Rimkutė, 2019) to extract

(19)

18 percentages of the processed words and percentages in the total words analysed. In order to be able to present and analyse trends, we have created a small database based on the results provided by this text mining in Wordstat. This database distinguishes between the publishing agency, the sort of report (annual report or work programme), the year of publishing, the relative dictionary percentage in relation to the cumulative dimensional dictionaries and the percentage of the dimensional dictionary relative to the processed text. We have chosen for this research design to create a comparable time series design in Stata which can be complemented by new agencies and publication forms.

For comparison between different agencies and the ESFS general trends, we will mainly look at the variable of keyword occurrence as a percentage of the total analysis. We expect that this percentage offers the most comparative insight when doing an analysis between an overall trend and the separate agencies. This percentage also gives us an overview of how likely agencies are to employ reputational behavioural strategies, as a higher cumulative percentage translates to a higher attention for these dimensions. In analysis of the results, we calculate a simple linear regression to predict the separate dimension based on publication year.

Within Wordstat, exclusions are made for numeric values and a list of standard English words not relating to any clear subject-matter, filtering surplus words and input not relating to any substantive content for analysis. Also applied in Wordstat is the option of stemming, reducing input words back to their base and thus increasing the accuracy in the analysis of our keywords.

(20)

19

4. Analysis

4.1. The general overview

With this analysis we line out development of reputational behaviour in the ESAs. Quantitative mapping of agency reputational behaviour to date has been limited to the broad collection of EU agencies and has seldom been done for a specific set of specialized agencies. Building further upon this exploration by Busuioc & Rimkutė (2019) and Rimkutė (2019), we run an explorative and quantitative analysis of specific financial, advisory and enforcing agencies housed in the ESFS.

First, we will show the cumulative development of the ESFS agencies overtime. This will provide initial insight of the agencies and a general trend within their development. The analysis on the ESFS as a whole is the point of departure for further comparative analysis. Our datapoints include an overview of communication measured by the average between annual reports and work programmes (fig. 1a. and 2a.) and a separate report of the annual reports and work programmes (resp. left and right in fig. 1b-c. and 2b-c.)

Within the further figures, we have presented the timeseries for each of the dimensions, split up per agency in order to show comparative trends within the figures presented. For our research, we have again made the distinction between a combined average of annual reports and the work programmes, as well as a separate trend for annual reports and work programmes respectively.

Figure 1a. Cumulative development of organizational reputation of the ESAs per year, in percentage of total words (all publications)

(21)

20

Figure 1b-c. Cumulative development of organizational reputation of the ESAs per year, in percentage of total words per publication sort (left: Annual Reports, right: Work Programmes)

Figure 2a. Cumulative development of organizational reputation of the ESAs per year, in percentage of analysed words (all publications)

Figure 2b. Cumulative development of organizational reputation of the ESAs per year, in percentage of total analysed words per publication sort (left: Annual Reports, right: Work Programmes)

(22)

21 This general overview shows an interesting initial focus on the moral dimension, next to the technical dimension for the ESAs. This focus on the moral dimension then shows a negative trend for the ESAs, as use of the moral dimensions declines from 2012 and onwards. The procedural dimension gains more attention in the second full year of operation, but diminishes overtime, creating an overall negative trend for this dimension. A possible explanation for this focus on the moral dimension could be the reasons behind the ESAs foundation. The ESAs were set up in wake of the financial crisis and their primary communicative focus may have been tailored to serve public interest and combat issues of distrust, thus focussing on the moral dimension. The moral dimension is initially prevalent in all of the separate agencies, we will further elaborate in the individual analysis.

Figure 3. Cumulative trends of organizational reputation of the ESAs per year. Average of the combined annual reports and work programmes, in percentage of total analysed words

2012 2013 2014 2015 2016 2017 2018

Technical 1,03% 1,13% 1,29% 1,18% 1,20% 1,32% 1,28%

Performative 0,64% 0,69% 0,70% 1,16% 1,03% 1,01% 0,89%

Procedural 0,67% 0,93% 0,86% 0,75% 0,74% 0,60% 0,61%

Moral 0,98% 0,97% 0,94% 0,96% 0,86% 0,79% 0,78%

Table 2. Average of use of reputational dimension in annual reports and work programmes, in percentage of total analysed words

Across time, the behaviour of the ESAs on average tends to vary heavily. While trends may say something for the field in which the agencies collectively operate within the ESFS framework, it is more interesting to analyse these agencies in their separate roles. The following paragraph is dedicated to the separate trends in reputational behaviour for EBA, EIOPA and ESMA.

(23)

22

4.2. ESAs trends

The analysis of individual agencies shows a more detailed overview of how reputational behaviour is employed by the separate ESAs. In order to create this overview, we present the overall analysis on the average use of reputational dimensions within publications per agency, and an analysis on the prospective annual reports and the retrospective work programmes separately per agency. The following figures show the development of reputational dimensions, for the EBA, EIOPA and ESMA.

Figure 4a-d. Average trends of organizational reputation per dimension through all publications, presented per agency in percentage of total analysed words

Coeff. Cons. F Sig. R²

Technical 0,04 -74,8 1,75 0,194 0,041

Performative 0,06 -124 4,46 0,041 0,100

Procedural -0,03 70,8 2,38 0,131 0,056

Moral -0,04 75,85 2,51 0,121 0,059

Table 3. Overview of simple linear regression of publication year on dimensions for the ESAs in general.

Overall the technical dimension is on average the most important for the ESAs. From 2012 through 2018 the technical dimension is most prevalent within the documents. This is in line with our expectation. Starting with 1,03% of total analysed words in 2012, the technical dimension grows with a coefficient of 0,04% per year (P<.2).

(24)

23 Contrary to our second expectations, the moral dimension is initially the most prevalent within the ESAs. With 0,98% of the total analysis in 2012, communication regarding trust, integrity and protection are initially important to the agencies in the ESFS framework. The attention for this dimension does decrease overtime with a coefficient of -0,04% per year (P<.2). While not meeting our expectation, the attention for the moral dimension can be explained. The nature of the ESFS foundations and their tasks in the wake of the financial crisis may have inclined the agencies to focus their communications on regaining the consumer’s trust. The breakdown of the individual agencies shows a similar trend for each of the ESAs.

While we cannot accept our second hypothesis, we should not fully reject it based on these results. The overall trend on reputational behaviour in the ESAs shows that the performative dimension is on the rise, especially in 2015 there is an increase in the performative dimension that holds. While initially starting at the lowest average of 0,64%, the procedural dimension holds the largest positive coefficient of 0,06% per year (P<.0.05). It does not overtake the technical dimension in the limited lifespan of the ESAs. It does however show a stronger growth than the technical dimension, if the enforcement role of the ESAs remains a primary role for the agencies, the performative dimension is likely to become the most emphasized dimension, in line with our expectations.

The procedural dimension is focused mainly in the foundation period for the agencies. While 2012 shows a relatively low attention for this dimension, the ESAs focus more on this dimension in 2013. Past 2013 the procedural dimension gradually decreases overtime, with an average coefficient of -0,03% per year (P<.2). This supports our expectation of young enforcing agencies focusing on procedural communication, diminishing overtime.

4.2.1. European Banking Authority

As can be seen in table 5a, the development of the EBA has an initial focus on the technical and moral dimension. With a respective 1,46% and 0,99% of their writing dedicated to these dimensions. Primary focus on the technical dimension remains over the analysed lifespan of the EBA. This expectation is in line with our first hypothesis, in which is stated that the technical dimension is most important for the ESAs agencies. The technical dimension remains the most important dimension for the EBA every year. The coefficient for the technical dimension is 0,04% per year, however, this result is not significant

The low attention for the performative dimension does not support our second hypothesis, in which we expect the supervisory nature of the EBA to incline them to focus on the performative

(25)

24 dimension – in 2015 they do shift their attention more towards the performative dimension, as their use of this dimension majorly increases from 0,57% to 1,25% of the total analysis. This shift is drastic and shows an increased focus on the performative dimension, however this growth does not seem to continue in 2016 and onwards as the dimensions remains relatively stable from 2015 and onward. Overtime the performative dimension has a positive coefficient of 0,11% per year (P<.2) over the publication years.

Due to no sufficiently significant results on both the growth of the technical and the performative dimension for the EBA, we cannot accept our third hypothesis. It cannot be proven that focus on performative reputation is becoming more prevalent than the technical dimension for the EBA. Our trends do show that use of the performative dimensions grows quicker than the technical dimension.

Initially, the procedural dimension has the lowest focus of all dimensions, but in 2013 the procedural dimension is the second most focused dimension. From 2013 through 2018 the focus on the procedural dimension declines. We cannot state a statistically significant coefficient for the procedural dimension. Also, we can not fully accept our fourth hypothesis, while the procedural dimension is the most focused for publication years 2013 and 2014, the hypothesis can’t be accepted due to the fact that it is the least focused dimension in the agency’s initial years.

The initial focus on the moral dimension is interesting, as this does not meet any stated expectations. The moral dimension starts as the second most important dimension and gradually decreases overtime, with a coefficient of -0,05% per year (P<.1).

Technical 0,04 -73,3 1,27 0,281 0,096

Performative 0,11 -211,4 2,07 0,176 0,147

Procedural 0,002 5,1 0 0,968 0

Moral -0,05 109,1 3,68 0,079 0,235

Table 4. Overview of simple linear regression of publication year on dimensions for EBA

2012 2013 2014 2015 2016 2017 2018

Technical 1,46% 1,44% 1,52% 1,50% 1,39% 1,74% 1,65%

Performative 0,64% 0,61% 0,57% 1,25% 1,05% 1,30% 1,00%

Procedural 0,46% 0,94% 0,93% 0,82% 0,65% 0,74% 0,67%

Moral 0,99% 0,88% 0,80% 0,67% 0,79% 0,54% 0,71%

Table 5a. Development of EBA’s use of reputational dimensions, measured by an average of annual reports and working programmes, over time in percentage of total analysed words

(26)

25 Our analysis tries to assess the effect of an external shock, in the form of Brexit, on the EBA’s behaviour. The referendum for Brexit was called mid-2016, on March 29, 2018 the UK formally notified the EU of its intention to invoke Article 50. Due to this time path, we expect to see a shift between the publication years 2016 and 2017 and a continuation thereof in later years. In 2017 we do indeed see a substantive increase in both the technical and the performative dimension. This is in line with our expectations, as output orientation drives the EBA towards an increased focus in these dimensions in order to re-establish themselves with their audiences. The technical dimension and the performative dimension peak for the EBA in 2017, especially in relation to EIOPA and ESMA. The procedural dimension shows a slight increase for EBA of 0,09% in 2017 and the moral dimension shows a large decrease of 0,25%.

What is also interesting is the differences between annual reports and working programmes within the EBA’s communications. As shown in Table 5b and 5c, the weight of the increased use of the technical and procedural dimension in 2017 comes mainly from the working programmes.

This can be easily explained due to the situation of Brexit, in which a deal or no-deal scenario is still unsure. Within their annual reports, seen in Table 5b, the EBA likely wants to emphasize its continuation and therefore does not shift dramatically in their use of reputational dimensions. Even in the annual reports there is an increased focus for the technical dimension. Attention for the performative dimension (and other dimensions) remains relatively low in the annual reports. The working programmes, as opposed to the annual reports, are prospective documents. Within communication through these documents, the EBA is likely to seek to assert its role on both the technical and performative dimension – or its output orientation in the wake of Brexit. The uncertainty surrounding the Brexit deal inclines them to use the prospective working programmes as main tool for communication surrounding the preparation of the EBA’s task for the process of the UK’s exit.

2012 2013 2014 2015 2016 2017 2018

Technical 1,78% 1,42% 1,36% 1,57% 1,27% 1,43% 1,39%

Performative 0,45% 0,46% 0,52% 0,62% 0,55% 0,57% 0,45%

Procedural 0,75% 1,42% 1,40% 1,16% 0,97% 0,83% 0,72%

Moral 0,72% 0,67% 0,79% 0,65% 0,62% 0,29% 0,59%

Table 5b. Development of EBA’s use of reputational dimensions, measured by annual reports, over time in percentage of total analysed words

(27)

26 2012 2013 2014 2015 2016 2017 2018 Technical 1,14% 1,46% 1,67% 1,43% 1,50% 2,05% 1,91% Performative 0,83% 0,75% 0,61% 1,88% 1,54% 2,03% 1,55% Procedural 0,16% 0,46% 0,46% 0,47% 0,33% 0,64% 0,61% Moral 1,25% 1,08% 0,81% 0,68% 0,96% 0,79% 0,83%

Table 5c. Development of EBA’s use of reputational dimensions, measured by work programmes, over time in percentage of total analysed words

4.2.2. European Insurance and Occupational Pensions Authority

Table 7 shows the development in reputational behaviour for EIOPA. In line with the overall trend and with EBA are the primary focus on the technical dimension, followed by the moral dimension. Initially these encapsulate 1,10% and 1,03% respectively of the total words analysed in the documents. The trends show a relatively stable development for the technical dimension, remaining the most important dimension with the exception of 2015, in which the moral dimension becomes most prevalent. Overall, the technical dimension holds a very small increase with a coefficient 0,01% per year. This result is not significant. Overall, the technical dimension is the most important dimension for EIOPA, supporting our first hypothesis.

EIOPA also shows a low attention for the performative dimension initially. Comparable with EBA and the general trend, EIOPA shows an increase in 2015. This continues up to 2016 and then again decreases in 2017 and 2018. Overall the performative dimension has coefficient of 0,03% per year. This result, however, is not significant. The lack of attention for the performative dimension is contrary to our expectations. The growth is bigger than that of the technical dimension in this case, but this cannot be statistically verified due to the results being insignificant.

The procedural dimension does not show any exceptional result, it does however decrease over time in line with our expectation. The procedural dimension shows a significant coefficient of -0,06% per year (P<.01).

EIOPA also shows an interesting unexpected focus on the moral dimension, more than any other agency they focus extensively on these communications. This is interesting as it an unlikely focus for an agency with advisory and enforcement roles. However, the nature of EIOPA’s scope, being insurances and pensions, may incline them to strategically position themselves towards consumers and their trust, thus focussing their communications on subjects as integrity, protection and values. The moral dimension shows a coefficient near 0, but this result is not significant. However, the trend does indicate a steady focus of the moral dimension for EIOPA.

(28)

27 In relation to our last hypothesis it is interesting to see that EIOPA is not showing abrupt changes in their reputational behaviour between 2016 and 2017. The performative dimension does decrease for this agency in 2017 and onwards. Theoretically, this cannot be linked to Brexit, as external shock for agencies is likely to increase the use of the technical and performative dimension. This may indicate that EIOPA is not directly affected by Brexit or that other factors are influencing its behaviour. EIOPA returns to their regular attention for the performative dimension, publication year 2015 and 2016 being outliers for this dimension in EIOPA.

Technical 0,01 -16,7 0,06 0,807 0,005

Performative 0,03 -55,9 0,66 0,431 0,052

Procedural -0,06 116,8 13,86 0,003 0,536

Moral -0,01 12,3 0,02 0,894 0,002

Table 6. Overview of simple linear regression of publication year on dimensions for EIOPA

2012 2013 2014 2015 2016 2017 2018

Technical 1,10% 1,17% 1,57% 1,34% 1,29% 1,17% 1,27%

Performative 0,75% 0,74% 0,79% 1,22% 1,20% 0,81% 0,83%

Procedural 0,72% 0,70% 0,65% 0,61% 0,60% 0,42% 0,39%

Moral 1,03% 0,98% 1,30% 1,59% 1,08% 1,10% 0,97%

Table 7. Development of EIOPA’s use of reputational dimensions, measured by an average of annual reports and working programmes, over time in percentage of total analysed words

4.2.3. European Securities and Markets Authority

In table 9 is the breakdown of ESMA’s reputational behaviour over time. Initially, ESMA appears to have a lowest interest for the sum of the dimensions in 2012. This only counts for the initial year, in 2013 ESMA moves closer to the sum of both the EBA and EIOPA.

An interesting finding, contrary to our first two expectations is the low interest for both the technical and the performative dimension in the communication of ESMA. With a percentage of 0,53% for both of the dimension, they make up for the lowest of the four. This is the case for 2012 and 2013. From 2014 and onwards the technical dimension overtakes the moral dimension. In 2015 the performative dimension becomes the most prevalent, followed by the procedural dimension against in 2016. From 2017 and onwards the technical dimension becomes the most prevalent dimension in ESMA’s communication. Due to this strong fluctuation within the dimensions, ESMA does not meet our first two expectations. Both the technical and the performative dimension show an upwards trend. The coefficient for the technical dimension is

(29)

28 0,07% per year (P<.05) and the performative dimension shows an insignificant coefficient of 0,05% per year. This does not meet our expectations in development of the dimensions. ESMA does show a downward trend from 2013 in the procedural dimension, meeting our expectation of an increased focus on this dimension in the younger period of enforcement agencies, with a diminishing attention in the following years. The procedural dimension shows a coefficient of -0,04% per year (P<.2).

Interestingly, ESMA also shows an initial preference for the moral dimension, which diminishes over time. The moral dimension starts off as the largest variable and then decreases with a coefficient of -0,05% per year (P<.2). With ESMA also having a strong focus on the moral dimension, all ESAs tend to do so contrary to our initial expectations.

Regarding our last hypothesis, ESMA does not show a big sudden shift in its reputational behaviour. In 2017 it does focus its communications more on the technical and performative dimension, but not with the same magnitude as EBA does. We can continue our support for our fifth hypothesis, as EBA shows the strongest shift related to both EIOPA and ESMA.

2015 shows a large increase for the performative dimension for ESMA, much like both the EBA and EIOPA. An explanation for this, is an increased focus on Key Performance Indicators or KPIs for short in line with our expectations.

Technical 0,07 -134,5 18,2 0,001 0,603

Performative 0,05 -105 1,66 0,222 0,122

Procedural -0,04 90,48 2,22 0,162 0,156

Moral -0,05 106,199 2,81 0,12 0,190

Table 8. Overview of simple linear regression of publication year on dimensions for EIOPA

2012 2013 2014 2015 2016 2017 2018

Technical 0,53% 0,77% 0,80% 0,71% 0,94% 1,06% 0,92%

Performative 0,53% 0,73% 0,76% 1,02% 0,86% 0,93% 0,85%

Procedural 0,85% 1,15% 1,00% 0,84% 0,99% 0,64% 0,77%

Moral 0,92% 1,06% 0,71% 0,63% 0,71% 0,74% 0,65%

Table 9. Development of ESMA’s use of reputational dimensions, measured by an average of annual reports and working programmes, over time in percentage of total analysed words

(30)

29

5. Discussion

By narrowing down our scope on a particular set of agencies, we attempted to deliver additional building blocks to the framework of organizational reputation. We have done so by analysing in detail how the financial and enforcement oriented ESFS agencies operate, look at their cumulative trends as ESFS, but also the agencies individual trends. How do these trends develop? Where can we find discrepancies in these trends and how can these be explained? What are the effects of an external shock? In our research, we have been able to pinpoint several expected and unexpected behaviours of these ESAs in the perspective of reputational behaviour.

We have researched five hypotheses in order to answer these questions and to test the existing framework of organizational reputation on a small scale, on specialized agencies. We have deducted the overall trends for the ESFS as a whole and also narrowed down our scope on the individual agencies. The trends and the development of the reputational behaviour have been held up against the expectations derived from the existing, limited framework of organizational reputation. Following in this chapter is a breakdown of the hypotheses, discussed on level of the ESFS and the individual agencies.

Due to the small size of our dataset, we cannot fully accept most of the hypotheses. The results of our trend analysis do however provide an exploratory insight in a more micro level research on European agencies and their reputational behaviour. It especially provides an insight and the reputational behaviour of the ESFS and its individual agencies of financial and enforcing nature.

Hypothesis 1: Technical reputation is most important for the ESAs. Their financial regulatory nature and their depoliticized nature incline them to focus more on the technical dimension over the other dimensions for the agencies.

For our first hypothesis, the general trend shows an overall primary focus of the ESAs on the technical dimension. In all publication years, 2012 through 2018, the cumulative average percentage of the technical dimension shows the highest results. The technical dimension also shows an upwards trend for all of the agencies, confirming their attention for this dimension. While there are some exceptions in individual cases, on average the technical dimension is the most emphasized dimension in each of the years. For EBA the technical dimension is most important for 2012 through 2018. This is also the case for EIOPA, with the exception of 2015 in which the moral dimension is most emphasized. ESMA is an exception, with the technical dimension only being the most emphasized in 2017 and 2018. The technical dimension does however show the strongest upward significant trend for ESMA.

(31)

30 While we cannot fully accept this hypothesis, the overall primary focus on the technical dimension is strong. We can support that adhering to the ‘gold standard’ and technical prowess is important for the ESAs (Majone 1998:16, Maor, 2007:963,974), also confirmed in the upward trend for all of the agencies. Focus on the technical baseline competences of agencies (Busuioc & Rimkutė, 2019:29) can be confirmed for the ESAs in this study. The upward trend, while small for EIOPA, is also visible for all agencies. Combined with high initial attention for this dimension, while we cannot accept this, we find our first hypothesis relevant.

Hypothesis 2: The ESAs focus on performative reputation second to technical reputation. Due to the nature of their supervisory/enforcement tasks in relation to national competent authorities, a strong focus on the performative dimension is necessary for the agencies.

Hypothesis 3: Over time, performative reputation will become more prevalent than technical reputation. The nature of the agencies will shift focus to its national competent authorities, as the ESAs are inclined to secure their roles relative to national competent authorities.

Regarding our second and third hypothesis, we have analysed the trend of the performative dimension in relation to the others. In this analysis, we have found an interesting result. Second to focus on technical dimension, the agencies did not have their focus on the performative dimension. Based on this we reject our second hypothesis, however from 2015 and onwards we see the performative dimension becoming the second most emphasized dimension in the cumulative average of the ESAs.

Each of the agencies show this peak in the performative dimension for this year, with a slight decrease following after 2015. A likely explanation for this is a possible EU commitment to orient more on performance-oriented results and KPIs. This argument is in line with Busuioc & Rimkutė (2019:29) in which a shifted focus on efficiency and output translates into a stronger emphasis on the performative dimension. Based on this research, we cannot deduct a reason for the shift for the ESAs in general. We do see that mainly the work programmes for EBA shows a drastic shift in its performative dimension. From publication year 2015 and onwards, the EBA has a visibly enhanced its reports, aimed on results and outputs (EBA, 2015). Due to this change, we cannot accept or reject our third hypothesis. While initially the performative dimension seems less important to the ESAs, there is a big upwards trend, mainly due to the changes in 2015. This hypothesis is still plausible, due to this visible trend break, and requires future research to see if this trend continues.

Developing Agencies and Their Organizational Reputation: A research of reputational behaviour of the supervisory agencies EBA, EIOPA and ESMA