• No results found

Perceived transparency in the financial world

N/A
N/A
Protected

Academic year: 2021

Share "Perceived transparency in the financial world"

Copied!
88
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

i Abstract

Master thesis

Corporate Communication

Perceived transparency in the financial world

Master: Corporate Communication, Communication Science Supervisor: W.J.L. Elving

Student: Redmar Poortstra (10047042) Date: June 24, 2014

(2)

ii Abstract

Purpose – Dutch banks have to regain its customer’s trust. The purpose of this study is to

present empirical evidence for the role of transparency and stakeholder engagement in the increase in trust and reputation. Additionally, it is the purpose of this article to present on which topics a bank should be more transparent.

Design/methodology/approach – Customers of Dutch banks were surveyed by the

means of an online questionnaire. The questionnaire measured the perceived

transparency, trust, reputation and stakeholder engagement. Additionally respondents were asked to name their bank and to give topics on which their bank should be more transparent.

Findings – The findings suggest a strong link between transparency and trust in the

organization. Additionally, it is concluded that transparency positively affects the

reputation of an organization. It appeared that stakeholder engagement did not strengthen the effects of transparency. Respondents indicated a need for more transparency on information that affects them directly, on information regarding their money and on information about the bank as an institution. Finally, this study concludes that there are differences in the levels of trust and reputation between banks.

Originality/value – This study adds to current literature by providing empirical findings

for the effects of perceived transparency on trust and reputation.

Keywords: Transparency, Trust, Reputation, Stakeholder Engagement, Bank,

(3)

iii

Master thesis

Corporate Communication

Perceived transparency in the financial world

Master: Corporate Communication, Communication Science Supervisor: W.J.L. Elving

Student: Redmar Poortstra (10047042) Date: June 20, 2014

(4)

iv

Table of Contents

List of Tables ... vi

List of Figures ... vii

Introduction ...1 Theoretical background ...5 Transparency ...5 Trust…. ...11 Reputation ...13 Stakeholder engagement ...14 Method ...18

Population and sample ...18

Data…. ...19 Variables ...21 Transparency ...21 Trust… ...22 Reputation ...23 Stakeholder engagement ...24 Ethics…...24 Results. ...25 Univariate descriptives...25 Hypotheses 1 and 3a ...26 Hypotheses 2 and 3b ...27

(5)

v Sub-question 1 ...30 Additional results ...35 Discussion ...38 Summary of findings...38 Managerial implications...42

Limitations and directions for future research ...43

Contributions...45

References ...45

Appendix A: Survey (in Dutch) ...54

Appendix B: PCA Transparency ...77

(6)

vi List of Tables

Table 1. Demographics of sample group versus Dutch residents………..20

Table 2. Rotated principal component analysis of trust……….23

Table 3. Means, standard deviations, skewness, correlations and reliability coefficients.26 Table 4. Regression models with transparency, Zmoderation, age, gender and education as predictors for trust and reputation……….30

Table 5. Means and standard deviations of trust across different banks………...36

Table 6. Bonferroni post-hoc test for banks and trust………...37

(7)

vii List of Figures

Figure 1. Transparency in different discourses………..6

Figure 2. The effect of stakeholder engagement and transparency on trust……….16

Figure 3. Conceptual model………..18

Figure 4. Conceptual model with beta’s………....29

(8)

Introduction

In today’s financial world, banks experience low levels of trust amongst their customers (Trust in financial services, 2014). This study explores the roles of

transparency and stakeholder engagement in increasing trust in banks. Moreover, the effect that transparency and stakeholder engagement have on the reputation of Dutch banks is examined in greater detail. After investigating these effects, it is examined which topics a bank could be more transparent about. Rawlins (2008) investigated the concept of transparency and concluded that transparency could be defined as “the deliberate attempt to make available all legally releasable information” (p. 75). The concept of stakeholder engagement is in this study defined as the involvement of stakeholders in corporate decisions and/or activities (Greenwood, 2007).

The examination of transparency has vaulted to prominence in recent years (Rawlins, 2008). Researchers across a wide variety of academic disciplines have used different methods to determine whether or not an organization is transparent, such as content analyses (Ball, Owen & Gray, 2000), case studies (Dubbink, Graafland &

Liedekerke, 2008) or interviews with organization’s top-management (Jahansoozi, 2006). In most cases however, the organization was the subject of analysis. Rawlins (2008) on the other hand examined organizational transparency from a stakeholder’s point of view. In the field of governmental transparency, it is more common to investigate the

perceptions of stakeholders, which are citizens in that case (e.g. Piotrowski & Ryzin, 2007). Rawlins’ (2008) study however, focuses on organizational transparency.

(9)

Rawlins’ (2008) units of analysis were employees of a hospital. In order to measure their perceptions, Rawlins constructed a transparency scale to measure the stakeholders’ perceptions of organizational transparency. In the current study, Rawlins’ original study from 2008 is partly replicated but with other units of analysis. In the final sentence of Rawlins’ article, the author states: “Providing a valid and reliable instrument that measures this complex concept (transparency red.) in such a way that provides an assessment by stakeholders of organizational transparency is the first step in answering some of these other important questions’’ (p. 97). In the current study, it is tested whether this statement is consistent with a group of stakeholders other than Rawlins’ employees, customers to be precise. Therewith this study attempts to contribute to the knowledge about organizational transparency by investigating the perceptions of organizational transparency amongst a new group of stakeholders.

Previous examinations of transparency led to several results. Dapko (2012) presented a connection between transparency, reduced skepticism, increased trust, purchase intention and attitude towards a firm. Rawlins (2008), Jahansoozi (2006) and Hon and Grunig (1999) also proposed a relationship between transparency and trust. Moreover, Rawlins (2008) concluded that employees rated their company’s reputation as more ethical when an organization is transparent, thereby presenting a link between transparency and ethical reputation. However Rawlins argued that further exploration of this link is required. This study investigates the relationship between transparency and trust again, although from a different viewpoint. Additionally my aim with this study is to

(10)

add to the current literature by further exploring Rawlins’ link between transparency and reputation.

Furthermore, this study attributes to current literature by the exploration of the role of stakeholder engagement in the relationship between transparency, trust and

reputation. Heretofore authors investigated the role of participation or dialogue within the concept of transparency (Jahansoozi, 2006; Rawlins, 2008). Dialogue is not exactly the same as stakeholder engagement; nevertheless, it could be seen as a form of stakeholder engagement (AccountAbility, 2005). Therefore it is investigated here whether;

stakeholder engagement is linked to the concept of transparency in the same way as dialogue is.

Another indication for the relevance of stakeholder engagement is the link it has with social auditing, a form of accounting with ties to social responsibility (e.g. Owen, Swift, Humphrey & Bowerman, 2000; Gao & Zhang, 2006). Although in a financial context, the authors presented transparency and stakeholder engagement as determinants for corporate social responsibility, thereby making clear a link between transparency and stakeholder engagement. In this study, this link will be further explored, by a survey amongst a bank its customers.

As stated earlier, the banking sector experiences low levels of trust nowadays (Trust in financial services, 2014). This study aims to provide an answer to the question of whether increased transparency and stakeholder engagement leads to a higher level of trust and reputation in the banking sector. Moreover, this study will provide Dutch banks with insights into the perceptions of their customers about the banks’ transparency

(11)

policies and the topics, which these banks could be more transparent on. We recommend these insights be used to stimulate the engagement of stakeholders in the decision making process of what information to make available. Furthermore, it might encourage banks to be more transparent on certain points. Therefore this study might well be of practical relevance to Dutch banks and their customers.

Against this background, the purpose of this research is to answer the following research question: What is the effect of transparency on trust and reputation and how is

this effect influenced by stakeholder engagement? The research question is accompanied

by a sub-question, which is: On which topics should a bank be more transparent? More specifically, this research has three objectives:

1. To provide empirical evidence which supports the role of transparency in increasing trust and the reputation of Dutch banks.

2. To explore the effect of stakeholder engagement on the above mentioned empirical evidence.

3. To provide insights into the perceptions of customers about which topics a bank should be more transparent about.

Customers of Dutch banks, preferably the four biggest banks – ING, Rabobank, ABN AMRO or SNS Bank (Bankensector-Nederland, 2014), are being surveyed to obtain these objectives.

This thesis has four parts, beginning with a review of the existing relevant literature. This is followed by a description of the research methodology and the data analysis. Next, the results of the four hypotheses and the sub-question are presented along

(12)

with two additional results. The thesis concludes with a discussion of the findings, managerial implications and recommendations for further research.

Theoretical background Transparency

According to the Oxford Dictionary (2014), transparency (of an organization) is defined as “open to public scrutiny”. However, the scientific concept of transparency is comprehensive and touched upon by many academic disciplines, for example: finance- (Liang, 2013), business- / organizational- (Meyer & Kirby, 2010), and ethical studies (Palanski, Kahai & Yammarino, 2011). Communication scholars also deal with the concept of transparency (e.g. Christensen, 2002). Nonetheless, this is often in combination with one of the three other academic disciplines. The communicational approach to the concept of transparency can be divided into three different levels of analysis. These are:

1. Macro level: investigating governmental transparency (Fairbanks, Plowman & Rawlins, 2007).

2. Meso level: examining organizational transparency (Wehmeier & Raaz, 2012). 3. Micro level: researching leadership transparency (Norman, Avolio & Luthans, 2010). This study will focus on the touch point between organizational and communication studies. The level of analysis in this study is meso level, investigating organizational transparency (see figure 1).

(13)

Figure 1. Transparency in different discourses

Heald (2006) divides transparency in four directions: two vertical ones and two horizontal ones. The vertical ones are on a micro level while the two horizontal ones are on a meso level and relevant to this study. The horizontal directions comprise

transparency inwards and transparency outwards. According to Heald, outwards transparency means that people from inside the organization can observe the outside world whilst transparency inwards means that people from outside the organization can observe what is going on inside the organization1. From Heald’s distinction it can be derived that organizational transparency is about the organization (inside) on the one hand and its stakeholder (outside) on the other hand.

A stakeholder is a person or a group who affects, or is affected by, the

organization (Freeman, 1984). If an organization decides to be transparent and send out

1

Christensen and Langer (2009) chose to turn these two around but it is chosen in this study to continue with the original division by Heald (2006).

(14)

information, the decision and the information will affect stakeholders. The organization is in this situation the sender of information whereas the stakeholder is the receiver. In organizational literature, this division into the organization on the one hand and the stakeholder on the other hand, is shown in the focus of a study. Some authors focus on the effects of organizational transparency by investigating the information that the organization disseminates (Ball, et al., 2000; Dubbink, et al., 2008). Other researchers focus on the stakeholders’ perception when examining the effects of organizational transparency (O’Neill, 2006; Rawlins, 2008).

The distinction between sender and receiver is also present in communication discourse. In the first half of the twentieth century, researchers investigated the power and influence of the mass media. The mass media was the sender and the unit of analysis in many studies (see Berger, 1995). It was suggested by many commentators that the receiver was a passive consumer of mass media, devoid of independent critique.

Notwithstanding, theories such as the magic bullet theory (Katz & Lazarsfeld, 1955) were proposed, which focused strongly on the sender of the message. Later on, these theories were falsified by studies such as “the effects of mass communication” by Joseph Klapper (1960). The communication discourse then began to shift towards a focus on the receiver of a message. This resulted, for example in agenda-setting theory (McCombs & Shaw, 1972; Carroll & McCombs, 2003). In contemporary communication discourse, the focus is still on the receiver of a message (Colombo, 2004).

The same division between sender and receiver can be found when reviewing communication literature about the concept of transparency. On the one hand, there are

(15)

authors who conceptualize transparency as the giving (Park & Hon, 2010) or sharing (Yang & Lim, 2009; Chen, Cheong & Li, 2010) of information. Thereby focusing on the organization (sender), which transmits information. While on the other hand there are communication scholars who acknowledge the need for a focus on the stakeholder (receiver) in research on organizational transparency (O’Neill, 2006; Christensen & Langer, 2009).

Hood (2006) also found these two different approaches but elaborated on the need to focus on the receiver of organizational transparency by stating that: “transparency is more often preached than practiced” (p. 3). Thereby insinuating that there is a difference between what the organization says it does (preach) and what it really does (practice). Investigating organizational transparency by analyzing an organization thus may result in an idealistic representation of the ‘preaching’. Whereas surveying stakeholders of an organization is more likely to uncover the reality of the organization’s transparency efforts, as external stakeholders have no interest in bending the truth. Therefore, in this study, organizational transparency is examined by surveying stakeholders’ perceptions. Thereby contributing to existing literature, which focuses on the receiver side of

organizational transparency (O’Neill, 2006; Rawlins, 2008; Christensen & Langer, 2009) and the current trend of contemporary communication discourse (Colombo, 2004).

Many scholars have proposed definitions of transparency without taking the stakeholder into consideration. For example, Liu and Horsley (2007) are communication scholars who defined transparency as honest communication. Bushman, Piotroski and Smith (2003) are organizational scholars who defined transparency as “the widespread

(16)

availability of firm-specific information concerning publicly listed firms in the economy to those outside the firm” (p. 210).

In both definitions, transparency is about the dissemination of information while the understanding of information is not taken into consideration. As argued above, in this thesis the understanding of information by the receiver is considered important. Therefore Gower’s (2006) definition of transparency suits better, which is that transparency is “a measure of the degree to which organizational actions and decisions are ascertainable and understandable by a party interested in those actions or decisions” (p. 95). In this

definition the receiver of the information is taken into account, thereby providing a more complete definition of transparency. In this respect, one might argue that the definition of Rawlins (2008) is even more accurate. According to Rawlins, transparency is “the

deliberate attempt to make available all legally releasable information –whether positive or negative in nature- in a manner that is accurate, timely, balanced and unequivocal, for the purpose of enhancing the reasoning ability of publics and holding organizations accountable for their actions, policies and practices” (p. 75). In Rawlins’ definition the stakeholder is taken into account just as in the definition of Gower (2006), however the contrast is that Rawlins (2008) adds to his definition; the purpose of the information (for enhancing the reasoning ability) and the way in which the information should be

transmitted (accurate, timely, balanced and unequivocal). Since Rawlins’ definition seems to be more comprehensive than the definition of Gower (2006), Rawlins' (2008) definition will be used for the remainder of this study.

(17)

Rawlins (2008) measures the transparency of an organization from a stakeholder’s perspective. According to Rawlins, there are four factors that underpin transparency:

1. The accountability of an organization 2. The openness of an organization

3. The rightness of the provided information, and

4. The participation of stakeholders in the decision process to decide which information should be disseminated.

Rawlins (2008) concluded that these four factors form a reliable measure. Therefore, in this study, Rawlins’ measure that includes the four dimensions is used to test and define organizational transparency.

Before moving on to the second central concept in this study, it is important to note that there is some ambiguity around the concept of transparency. On the one hand, there are authors who state that transparency can be regarded as ‘information disclosure’ (e.g. Chen, Cheong & Li, 2010; Park & Hon, 2010). While on the other hand, researchers argue that disclosure is a simplistic form of transparency, without taking the needs of the stakeholders into consideration (Rawlins, 2008; Coombs & Holladay, 2013). Moreover, a third group of researchers use the two concepts intertwined, thereby insinuating they can be perceived of as the same (Cheung, Jiang & Tan, 2010). It is beyond the scope of this paper to join the discussion about transparency versus disclosure. In this paper,

information disclosure will be regarded as part of transparency. Therewith aligning with the definition of transparency by Rawlins (2008).

(18)

Trust

Next to transparency, the second central concept in this study is trust. Morgan and Hunt (1994) advocated that trust occurs when a party has confidence in the other party’s reliability and integrity. In the context of this study, it means that the stakeholder of a company has confidence in the organization’s reliability and integrity. Despite the fact that the definition of trust by Morgan and Hunt (1994) is 20 years old, it is still cited in many contemporary studies (e.g. Vahlne, Schweizer & Johanson, 2012; Arnett & Wittmann, 2014; Giannakis & Harker, 2014). Therefore Morgan and Hunt’s (1994) definition of trust constitutes the basis of the definition of trust in this study, which is: the confidence of a stakeholder in the organization’s reliability and integrity.

Now that the concept of transparency and the concept of trust are defined in the context of this study, the relation between the two will be explicated. Many authors have linked transparency to trust, although research often fails to empirically support this claim. For instance, Bentele and Seidenglanz (2008) report that transparency is a factor leading to trust. The authors refer to the previous work of Bentele (1994) to support their claim. However, the latter article is purely normative. Jahansoozi (2006) presented empirical evidence for the relationship between trust and transparency. In Jahansoozi’s interview study, one respondent stated about transparency: “it helps build trust and it helps ensure that the community knows exactly where you stand’’ (p. 950). Nonetheless, this empirical finding is derived from a qualitative interview, making it difficult to generalize to any sample other than that participant. The reliability could also be

(19)

questioned, as the respondent was an industry member and thus a sender and not a receiver of a message.

Rawlins (2008) did investigate receivers of a message. Rawlins found that transparency was closely linked to integrity but that integrity in turn was more closely linked to trust. Rawlins comments:

"Integrity also seems to be important, but may be more broadly related to the concept of trust than the narrower concept of transparency. The relationship … suggests that they are closely related, and may provide some evidence that trust and transparency are linked” (p. 95).

In this respect, there might be empirical evidence for a correlation between trust and transparency but not causation. Moreover, Rawlins investigated employees and not other stakeholders. Furthermore, it could be argued that employees’ objectivity might be difficult to assess, since they have a special interest in the company.

Altogether, there is no conclusive evidence found in communication or organization literature for the relationship between trust and transparency from a stakeholder’s point of view. Nonetheless, some authors have proposed a relationship between trust and transparency (Jahansoozi, 2006; Rawlins, 2008). This relationship may partly be validated by literature from other academic disciplines, such as empirical findings in the public administration discourse by Kim and Lee (2012), who advocated that participants’ trust in the government correlated positively with perceived

(20)

organization literature, and empirical evidence from another academic discourse do lead to suspect that:

H1: There is a positive correlation between the perceived transparency of and the trust in

an organization.

Reputation

In addition to trust, reputation is a concept widely linked to transparency. But before this link is explained, first the concept of reputation will be discussed. According to van Riel and Fombrun (2007) “reputations are overall assessments of organizations by their stakeholders.” (p. 43). The reputation of a bank for example consists of the overall assessment of the bank by the bank’s stakeholders. This means that the only way to assess a bank’s reputation is to ask its stakeholders about what they think of the bank. There are many more definitions of reputation (see Barnett, Jermier & Lafferty, 2006: van Riel & Fombrun, 2007). However to give a complete overview is beyond the scope of this study. In this study the definition of reputation by van Riel and Fombrun (2007) is considered adequate.

Van Riel and Fombrun (2007) additionally presented six communication traits that correlate with high reputations. One of these traits is transparency. The researchers came to this conclusion by analyzing seven years of data from research all over the world. One slight remark about this research is that the researchers talk about

transparency and openness as one of the same. However, according to the definition by Rawlins (2008), transparency is more than just being open. Transparency is also being

(21)

accurate, timely, balanced and unequivocal. Van Riel and Fombrun (2007) do not take these characteristics of transparency into account. Nonetheless, a preliminary indication of a link between transparency and reputation has been established and proven.

Likewise Rawlins (2008) investigated the link between transparency and reputation. However, the researcher only examined a small part of the company’s reputation. Rawlins concluded that employees perceive their employer as having a more ethical reputation after being transparent. Therewith claiming that there is a positive link between transparency and the ethical part of a reputation. Taking this finding, together with the findings by van Riel and Fombrun (2007), it is proposed that transparency has a positive effect on reputation. The hypothesis that will be tested is that:

H2: Perceived transparency positively affects the organization’s reputation.

Stakeholder engagement

The fourth and final central concept in this study is stakeholder engagement. Stakeholders are people or groups of people who affect the organization’s goals, or who get affected by them (Freeman, 1984). When a company makes an effort to involve stakeholders in organizational activities or decisions, it can be understood as stakeholder engagement (Greenwood, 2007). There are plenty organizational activities. Similar, there are many forms of stakeholder engagement. For instance, there is stakeholder

engagement as a form of: corporate governance (Freeman, 1984; Van Buren, 2001), legitimization (Deegan, 2002), accountability (Gray, 2002) or participation (Arnstein, 1969). The latter two are of particular interest to this study.

(22)

According to Gray (2002), stakeholder engagement can be understood as a form of accountability. This means that through the involvement of stakeholders in an

organization’s action or decision, the organization can be held accountable when it fails to meet its promises. This principle is somewhat similar to the accountability dimension of transparency as described by Rawlins (2008). According to Rawlins transparent organizations are accountable because stakeholders have the possibility to evaluate the company’s decisions. Additionally, stakeholder engagement as a form of participation (Arnstein, 1969) has similarities with the participation factor of transparency (Rawlins, 2008).

Next to these two links between stakeholder engagement and transparency, there is a third connection between the two concepts. This connection is made clear on the basis of an example. Hooks, Coy and Davey (2002) investigated the information

disclosure amongst companies and the stakeholders’ information needs. The researchers concluded that there is an information gap between the two parties. To overcome this gap, the researchers suggest that companies should organize stakeholder panels to listen to the stakeholders’ needs. The stakeholder panels can be understood as a form of stakeholder engagement. The stakeholders are engaged in the decision to which

information will be disclosed. As argued before, information disclosure might in turn be interpreted as transparency, thus the example by Hooks, et al. (2002) illustrates that stakeholder engagement strengthens the effects of transparency.

Stakeholder engagement not only influences transparency, it also affects the relationship between transparency and trust (Jahansoozi, 2006). Trust will increase when

(23)

an organization becomes transparent and will increase even more when the organization engages stakeholders in decision-making processes. This is illustrated with an example about Royal Dutch Shell. Shell suffered from a decrease in trust after a crisis. Shell started to regain trust by being transparent. Only when stakeholders were consulted (stakeholder engagement) about solutions and could participate in the decision-making process, did the trust in Shell began to increase heavily. Figure 2 shows the increase of trust in Shell after engaging stakeholders.

Figure 2. The effect of stakeholder engagement and transparency on trust (adapted from

Jahansoozi, 2006, p.944)

Furthermore, Jahansoozi (2006) concluded that trust increases drastically when transparency is accompanied by collaboration with stakeholders. Collaboration with stakeholders can be regarded as stakeholder engagement (Greenwood, 2007). Therefore, it might be concluded that Jahansoozi (2006) demonstrated empirical evidence for the effect of stakeholder engagement on the relationship between transparency and trust.

(24)

However, Jahansoozi’s findings are not generalizable due to the qualitative nature of the research method.

Nevertheless, the findings of Jahansoozi (2006) are supported by García-Marzá (2005), who on the basis of discourse ethics (Habermas, 1989) found that companies should be trusted if they meet ten criteria. Two of these criteria are dialogue and transparency. When an organization acts according to those criteria (i.e. when it is transparent), it can expect higher levels of trust. The research by García-Marzá indicates that there might be a relation between the concepts of trust, transparency and dialogue. As stated earlier, dialogue can be perceived as a form of stakeholder engagement

(AccountAbility, 2005). Therefore it can be concluded that there is a normative argument (García-Marzá, 2005) and empirical proof (Jahansoozi, 2006) for the positive effect of stakeholder engagement on the relationship between transparency and trust. Additionally, stakeholder engagement enforces the effects of transparency (Arnstein, 1969; Gray, 2002; Hooks, Coy & Davey, 2002). It is, therefore, posited that:

H3a: The positive relationship between transparency and trust gets stronger when

stakeholders are engaged.

There is no empirical evidence for the influence of stakeholder engagement on the relationship between transparency and reputation. Nonetheless, based on the argument that stakeholder engagement enforces the effects of transparency (Arnstein, 1969; Gray, 2002; Hooks, et al., 2002), it is hypothesized that:

H3b: The positive effect of transparency on reputation gets stronger when stakeholders are

(25)

Figure 3 represents the four hypotheses in one conceptual model.

Figure 3. Conceptual model

The main focus of this study is on the concept of transparency. Therefore after examining the effects of transparency, it might be of interest to explore the topics of transparency that the banks’ customers wish for. The exploration of the customer’s perceptions gives more depth to the quantitative results from the previous hypotheses. The sub-question that will direct this exploration is:

SQ1: On which topics should a bank be transparent?

Method

Population and sample

This survey study took place in the Dutch banking sector. Customers of Dutch banks were questioned. In theory, all Dutch residents may be customer at one of the banks. Therefore, demographics such as age, education and gender of the sample group should preferably be consistent with demographics of the Dutch population. To be able to compare the sample with the Dutch population, data from the statistical bureau of the

(26)

Dutch government (CBS) was used. For the division of the bank’s customers, data provided by Bankensector-Nederland (2014) was used. Table 1 shows that the division between the banks in the sample is almost the same as in Dutch society. Most

respondents were customer of the ING, followed by Rabobank, ABN AMRO and SNS Bank.

Other demographics of the sample group also showed many similarities with the majority of Dutch residents. The average age of the sample group was only 5 years older than the average Dutch resident and the male/female distribution was virtually the same. The sample group however, was better educated than the average Dutch residents (see Table 1 for specific statistics).

Data

Respondents of the online survey were non-randomly approached through e-mail and social media (Twitter and Facebook). I used my personal network to approach respondents, and used a snowball sampling technique (Biernacki & Waldorf, 1981). This might explain the higher level of education of the people in the sample group than the average educational level of the Dutch population. The survey was in Dutch because of the context of this study, which was the Dutch banking sector.

People who were not a customer of a bank could not participate in this study. In total 175 people started the survey but only 125 completed the survey (N=125). The relatively high dropout rate of 27% might be explained by the opt-in design of this study. People were free to participate but also free to quit before the end. A lack of incentives

(27)

might be another explanation of the high drop out rate. The youngest participant was 19 years old while the oldest respondent was 75 years old.

Table 1

Demographics and division across banks of sample group versus Dutch population

Banks

Sample group Dutch population

N % % ING 48 35,8 40 Rabobank 43 32,1 34 ABN AMRO 27 20,1 19 SNS Bank Other 9 7 6,7 5,2 4 3 Demographics Male 60 48 49,5 Female Age 65 52 50,5 M SD M 44,5 16,6 39

Education HBO/WO - MBO/HBO

Note. Statistics about Dutch Residents: CBS (2014). Retrieved May 14, 2014 from

http://www.cbs.nl/nl-NL/menu/cijfers/kerncijfers/default.htm. Bankensector-Nederland (2014). Retrieved April 10, 2014 from http://www.banken.nl/bankensector/bankensector-nederland.

The questionnaire (see appendix A) started by informing the respondent about the topic and goals of the study. Furthermore, fully informed consent was obtained from respondent’s to use their answers for scientific purposes. Without this consent, the

respondent couldn’t proceed with the questionnaire. The opportunity to ask questions was given and it was stated that neither a bank nor any party other than the researcher,

commissioned the research. The survey then continued with the instructions and questions. All questions were closed-ended except for the question about the age of the

(28)

respondents and the question “On what topics should your bank be more transparent?” The latter question had five text fields in which the respondents could give open answers about subjects that their bank should be more transparent on. This data was used for the first sub-question.

Variables

In this study, there were four main variables: transparency, trust, reputation and

stakeholder engagement. Below these will be described in detail.

Transparency

Rawlins’ (2008) multidimensional transparency scale underpinned the assessment of the respondent’s perception of a bank’s transparency. This 28-item scale was

translated into Dutch. One item was excluded because after translation, two items became nearly the same. An example of an item is: “My bank gives accurate information”. Respondents had to indicate on 7-pointsscales (Likert; 1 (Totally disagree) to 7 (Totally agree)), to what extend they agreed with the statements.

Based on the Eigenvalues, the rotated principal component analysis (PCA) showed that the 27 remaining items measured six dimensions of transparency. This differs from the original transparency scale (Rawlins, 2008), which had four dimensions. The reason for this difference may be due to the translation of the items in Dutch.

Nonetheless, the scree plot shows a clear inflection point at the fifth dimension, therefore only the first four dimensions are considered as important. These four dimensions explain

(29)

61.51% of the variance in transparency. However, as discussed in the theoretical background section, the definition of transparency includes all four dimensions: substantial information, participation, accountability and openness. All dimensions should thus be taken together to form one variable. In that respect, it was chosen to stay as close as possible to the original scale by Rawlins (2008) and to include all 27 items. Moreover, in this study the dimensions are of no importance to the research question. Therefore it is chosen to combine them all into one transparency measurement (see appendix B for the PCA table and Eigenvalues).

A reliability analysis of the 27-item transparency scale indicated a Cronbach’s alpha coefficient of 0.94, which suggests good internal consistency reliability. The Cronbach’s alpha could not be improved by deleting an item. A higher score on the measurement indicates a higher perceived transparency.

Trust

The respondents’ trust in a bank was assessed through a one-dimensional scale consisting of five items (see table 2). Four items were based on Morgan and Hunt’s (1994) existing scale while one item was derived from Rawlins’ (2008) trust scale. The added item was: “My bank is honest”. All items were measured on a seven-point Likert scale with scores ranging from 1 (Totally disagree) to 7 (Totally agree). The Cronbach’s alpha appeared to be good (α=.94). The scores of the five items were averaged to provide an overall trust score. Respondents with a higher score had more trust in their banks.

(30)

Table 2

Rotated principal component analysis of trust

Component 1

My bank is reliable .88

My bank is fair .91

My bank speaks the truth .90

My bank does the right thing My bank is honest

.88 .90

Note. Eigenvalue = 3.98, explained variance = 79,50%

Reputation

The reputation of a bank was measured using a multidimensional scale, consisting of 16 items. The customer-based corporate reputation (CBR) scale by Walsh and Beatty (2007) formed the basis of the reputation scale in this study. However, items concerning the organization as an employer were excluded. Additionally, five questions were deleted because of the low correlations with the dimensions of the CBR scale. The PCA showed that the reputation scale in this study consists of three dimensions, but based on the scree plot and the Eigenvalues it was chosen to force the items into one dimension. That dimension had an Eigenvalue of 8.23 and a total variance explained of 50.4% (see appendix C).

All scale points were labeled ranging from 1 (Totally disagree) to 7 (Totally agree). An example of an item includes: “My bank is a strong company”. There were no reversed items. The Cronbach’s alpha of the reputation scale was considered good (α= .94). The Cronbach’s alpha could not be improved by deleting an item. The responses to

(31)

the 16 items were averaged to determine an overall reputation score. A higher overall score indicated a higher reputation of the bank.

Stakeholder engagement

In order to assess the degree to which respondents felt their bank was engaging them in decisions and actions, a one-dimensional 10-item scale was used. Brunig, Dials and Shirka’s (2007) scale formed the basis for the scale in this study. Although Brunig et al. (2007) used their scale in the context of cities, the scale proved to be reliable for organizations as well. The Cronbach’s alpha was .93. The respondents’ scores were averaged to determine an overall stakeholder engagement score. A higher overall score indicated that respondents perceived that their bank had a high level of stakeholder engagement.

Ethics

Three precautions were made to ensure that the respondents were treated ethically. First of all, respondents had to express fully informed consent. Second, the respondent was given the opportunity to contact the researcher about questions regarding the research. However, it turned out nobody used this opportunity. Finally, it was

specifically mentioned that their answers would be completely anonymous and only for the purpose of this study. This was especially important because questions about a person’s bank might lead to suspicion.

(32)

Results

Univariate descriptives

Table 3 contains the means, standard deviations, skewness, correlations and Cronbach’s alpha reliability coefficients of the constructs measured in this study. The means for transparency, trust and reputation are all a little more positive than the middle of the scale while the mean of the stakeholder engagement score is a little more negative. The distribution of Trust is moderately skewed left. Transparency, stakeholder

engagement and reputation are distributed approximately symmetrically. All variables

are within the acceptable norms of skewness (Bulmer, 1979), so no adjustments have to be made.

Additionally, it appears that all variables correlate positively with each other. This indicates that changes in the scores of one of the four variables will positively affect one of the other variables. For example, there is a significant, strong positive correlation between the perceived transparency by a respondent and the score on the trust scale. The more one perceives his or her bank to be transparent, the more the respondent trusts his or her bank, r = .64, p < .001. All other variables have also a significant, strong positive correlation with each other. Except for the correlation between trust and stakeholder engagement, which have a significant, but moderate positive correlation with each other,

(33)

Table 3

Means, standard deviations, skewness, correlations and reliability coefficients (N=125)

Mean SD Skewness Transparency Trust Reputation Stakeh. Eng.

Transparency 4.56 0.83 -0.37 (.94) .64* .71* .67*

Trust 4.96 1.12 -0.76 (.94) .64* .46*

Reputation 4.72 0.88 0.04 (.94) .73*

Stakeh. Eng. 3.80 1.16 -0.04 (.93)

Note. Reliability coefficients in parenthesis; *= p < .001 (2-tailed)

Hypotheses 1 and 3a

The first hypothesis stated that there is a positive correlation between the

perceived transparency of, and the trust in an organization. Whereas hypothesis 3b stated that the positive relationship between transparency and trust gets stronger when

stakeholders are engaged. Both hypotheses can be tested in a single model. Therefore, the two hypotheses are presented along in the following paragraph.

In order to test these hypotheses, a multiple regression analysis is conducted. In the regression model, trust is the dependent variable and transparency, age, gender and

education are independent variables. Additionally, the multiplication of the z-scores of transparency and stakeholder engagement is also included as independent variable. This

new variable is called Zmoderation and assesses the moderation effect of stakeholder engagement on the relationship between transparency and trust (hypothesis 3b). Since a multiple regression analysis is conducted the results below rest on the assumption that if an effect changes, the other independent variables in the model are held constant.

The regression model is significant, F (6,111) = 13.39, p < .001. Because of the significance, the regression model can be used to predict trust. The strength of this

(34)

prediction is moderate, 43.3% of the variation in trust can be predicted on the basis of the independent variables, listed above (R2 = .43). Transparency has a significant, strong association with trust b* = 0.62, t = 6.21, p < .001, 95% CI [0.58, 1.12]. One point higher score on the transparency scale, the trust in the respondent’s bank increases by 0.85 points. This result thus confirms hypothesis 1. There is an effect relationship between the perceived transparency of and the trust in an organization.

However, the moderating variable Zmoderation appears to have no significant association with trust, b* = 0.05, t = 0.62, ns, 95% CI [-0.11, 0.21]. This means that the score on trust will not significantly change if scores on stakeholder engagement vary. Therefore, hypothesis 3b cannot be supported with the current sample. The positive relationship between transparency and trust is not strengthened when stakeholders are engaged. Moreover, both findings were controlled for age, gender and education. But neither age, b* = 0.07, t = 0.87, ns, 95% CI [0.01, 0.02] nor gender, b* = 0.01, t = -0.12, ns, 95% CI [-0.37, 0.32] or education, b* = -0.09, t = -1.09, ns, 95% CI [-0.26, 0.08] has a significant relationship with trust. So the findings of the regression model will not change if the score on age, gender or education vary.

Hypotheses 2 and 3b

Hypothesis 2 stated that the perceived transparency positively affects the organization’s reputation. Additionally, hypothesis 3b stated that the positive effect of transparency on reputation gets stronger when stakeholders are engaged. Again, both hypotheses can be tested in one model.

(35)

In the multiple regression analysis, reputation is the dependent variable.

Transparency, Zmoderation, age, gender and education are the independent variables.

The regression model is significant, F (6,111) =32.46, p < .001 and can be used to predict

reputation. This prediction is strong, transparency, Zmoderation, age, gender and education predict 65% of the variance in reputation (R2 = .65). Transparency has a significant, moderately strong association with reputation, b* = 0.42, t = 5.35, p < .001, 95% CI [0.28, 0.61]. The respondents’ score on reputation increases with 0.45 points if his or her score on transparency increases with 1 point and all other variables are held constant.

This finding is in itself not enough to support hypothesis 2. The regression analysis only tests if there is an association. In order to make a statement about the causality of the association, an additional question was posed. It appears that 72.6% of the respondents would give their bank a higher score on reputation if the bank would be more transparent. This finding together with the above-mentioned result makes it safe to assume that there is a causal relationship between transparency and reputation. Thus, the perceived transparency positively affects the organization’s reputation. Hypothesis 2 is thereby supported.

Nonetheless, this effect does not get stronger when stakeholders are engaged.

Zmoderation has no significant association with reputation, b* = 0.03, t = 0.41, ns, 95%

CI [-0.08, 0.18]. Hypothesis 3b has to be rejected. The positive effect of transparency on

reputation does not get stronger when stakeholders are engaged. Additionally, the control

(36)

95% CI [-0.01, 0.00], gender, b* = -0.08, t = -1.29, ns, 95% CI [-0.35, 0.07] and

education, b* = -0.07, t = -1.21, ns, 95% CI [-0.17, 0.04] have no significant influence on reputation.

In summation, hypotheses 1 and 2 are supported; transparency has a positive association with trust and a positive effect on reputation. However, there appears to be no moderation effect of stakeholder engagement on either one of the relationships, so

hypotheses 3a and 3b are rejected (see figure 4). All results are controlled for age, gender and education (see table 4).

(37)

Table 4

Regression models with transparency, Zmoderation, age, gender and education as predictors for trust and reputation

Trust b* Reputation b* Transparency 0.62* 0.42* Zmoderation Age Gender Education R2 F 0.05 0.07 -0.01 -0.08 .43 13.39* 0.03 -0.09 -0.08 -0.07 .63 32.46* Note. N= 112; *= p < .001 Sub-question 1

The first sub-question focuses on the information that banks should be more transparent about. Data was gathered through the open question “on which topics should your bank be more transparent?” The qualitative data was analysed with a grounded theory approach (Glaser & Strauss, 1967). Answers where read multiple times until broad themes could be formed through inductive coding.

Three broad themes emerged from the data. One theme indicated all information that was relevant to the customer. The following theme was about money-oriented information and the last theme focused on information about the bank. Each of these three themes is composed of sub-themes (see figure 5).

The first theme was customer-oriented information. Four sub-themes

underpinned this category. The sub-themes are all related to information that affects the customer. The first sub-theme is additional explanation. Respondents would like to have more or better explanations when the bank is communicating to them. A respondent’s

(38)

answer from this category is for example: “explanation of the contract’s small prints and consequences for the customer”. When a change affects the customer, the bank should be transparent and explain the differences and consequences. Additionally the bank should be clearer on the terms and conditions in contracts.

The second sub-theme is information on interest rates. Respondents indicated that they wanted more transparency from the bank when it came to interest rates and the changes thereof. When asked about subjects of transparency, one illustrating answer is: “decrease/increase in interest rates”. Apparently customers would like to receive information when a change in interest rates affects them. Therefore, this sub-theme fits into the broader theme of customer-oriented information.

The third sub-theme is risk of bankruptcy. Respondents indicated that they would like their bank to be more transparent about the risk of bankruptcy for the bank. The respondents stated for example that they needed more transparency on issues such as: “financial status” or “how the bank covers for bankruptcy”. Risk of bankruptcy is part of the larger theme customer-oriented information since the bank’s bankruptcy will affect all customers.

The final sub-theme of the customer-oriented information theme is privacy. Respondents indicated that they wanted more transparency on the subject of privacy because they wanted to know what the bank does with their personal data. An answer from this category is: “What do they [the bank red.] do exactly with my payment data?” What happens with a customer’s personal data is relevant to the customer and therefore part of the customer-oriented information theme.

(39)
(40)

The second theme was money-oriented information. The theme consists of two underlying sub-themes. The first sub-theme is information about investments.

Respondents wished to receive more information on the type of investment, for example a respondent answered: “more information on how the bank invests”. Additionally the respondents wanted more information on the companies, which the bank invested in. For example, “Invest in green, sustainability, no weapons” or “where do the banks invest in (sustainable causes)”. These answers indicate that respondent feel that they should be more informed about what happens to the money they deposit in the bank.

The second sub-theme of money-oriented information is somewhat similar to the previous sub-theme. It is labelled as what happens with money. Respondents indicated that they wanted to know what happened with their money. One respondent’s answer illustrated this: “It is my money, what happens with it?” This shows that respondents feel that the money they deposit in the bank is still theirs and that they need information on what the bank does with it. Because money is the central subject in this sub-theme, it is part of the broader money-oriented information theme.

The third theme is bank-oriented information. The theme consists of four sub-themes. The sub-themes are dealing with the bank as the central subject of information. The first sub-theme is about issues and faults of the bank and is labelled crisis.

Respondents indicated that the bank should be more transparent when it is in crisis whereas after the crisis, the respondents want the bank to be more transparent about potential solutions and expect information on actions to prevent the next crisis. An

(41)

answer to exemplify the crisis sub-theme is: “addressing affairs and prevent new scandals”.

The second sub-theme is compensation and bonuses. The respondents wanted to be better informed about the bonuses that the bank’s top management receive. Furthermore the respondents wished for their bank to be more transparent about salaries of the bank’s employees in general. The answer “bonuses” was given multiple times, other answers include “top salaries” and “salary of top management”. Both crises and bonuses deal with the bank as the central subject, therefore they are grouped within the broader theme of

bank-oriented information.

The third sub-theme of the bank-oriented information is information about the bank as an employer. Respondents indicated that they needed their bank to be more transparent about “labour conditions” and “retirement plans”. Apparently the way the bank treats its employees is also important to the customers. The final sub-theme of

bank-oriented information is business conduct. In this sub-theme, information about the bank’s

mission, vision, strategy and year figures is covered. Respondents gave answers, such as: “year figures”, “what is their [the bank’s red.] vision” and “profit/losses”. The answers indicate that the bank should be more transparent or a better communicator about topics regarding their business.

Altogether, the respondents indicated several topics on which their bank should be more transparent. The topics are categorized in three broad themes customer-, money-

and bank oriented information. The three themes are constituted of ten sub-themes that

(42)

bank should be more transparent about? The answer is the list of sub-themes, which are:

additional explanation, interest rates, risk of bankruptcy, privacy, investments, what happens with money, crisis, bonuses, employer and business conduct.

Additional results

Although not of direct interest to the research question, it was also tested whether

stakeholder engagement has a direct association with reputation or trust. The regression

model with trust as dependent variable and stakeholder engagement, age, gender and

education as independent variables is significant, F (4,111) =7.79, p < .001. Stakeholder engagement, age, gender and education predict 22.6% of the variance in reputation (R2 = .23) this is a weak predictor. Stakeholder engagement has a significant positive

association with trust, b* = 0.47, t = 5.42, p < .001, 95% CI [0.29, 0.63].

Additionally the regression model with reputation as the dependent variable and

stakeholder engagement, age, gender and education as independent variables is also

significant, F (4,111) =33.24, p < .001. The independent variables have a significant, strong association with reputation (R2 = .55). Stakeholder engagement has a positive association with reputation, b* = 0.72, t = 10.99, p < .001, 95% CI [0.45, 0.65]. One point higher on the scale of stakeholder engagement indicates a .46 increase in trust and a .55 increase in reputation. The associations of stakeholder engagement with trust and

reputation are controlled for age, gender and education. None of these yielded significant

(43)

The second additional result that is found in the data is the difference in the levels of trust and reputation between customers of Dutch banks. In the survey, respondents were asked to name their bank. It was then tested whether there are differences in the scores on trust and reputation between customers of different banks (See table 5).

Table 5

Means and standard deviations of trust and reputation across different banks

N M SD Trust ING 47 4.62 1.14 Rabobank 40 4.98 1.07 ABN AMRO 26 5.30 0.82 SNS Bank Other 9 7 4.56 6.37 1.30 0.45 Total 129 4.96 1.12 Reputation ING 45 4.42 0.80 Rabobank 39 4.91 0.69 ABN AMRO 25 4.76 0.88 SNS Bank 9 4.07 0.72 Other 7 6.27 0.33 Total 125 4.72 0.88

Note. Respondent are grouped in category “Other” when they are customer of another

bank than the four biggest Dutch banks: ING, Rabobank, ABN AMRO or SNS Bank.

Firstly, a one-way analysis of variance (ANOVA) was carried out to assess the influence of the bank where a respondent is a customer (bank) on trust. A significant, weak effect of bank on trust was found, F (4, 124) = 5.44, p < .001, η2 = .149. A

Bonferroni post-hoc test indicated significant differences between respondents of another bank than ING, Rabobank, ABN AMRO or SNS Bank (Other) and respondents who were customer of ING, Rabobank or SNS Bank (see table 6). For example, respondents who

(44)

were customers of the ING trusted their bank on average 1.75 points less than respondents who answered “Other”.

Table 6

Bonferroni post-hoc test for banks and trust (only significant differences)

Mdifference p

Trust

ING * Other -1.75 .001

Rabobank * Other -1.40 .015

SNS Bank * Other -1.82 .008

Note. Significant at the .05 level

Additionally, a one-way analysis of variance was carried out to assess the

influence of bank on reputation. A significant, weak effect was found, F (4, 120) = 11.30,

p < .001, η2 = .274. The Bonferroni post-hoc test indicated significant differences between ING and Rabobank, ING and Other, Rabobank and SNS Bank, Rabobank and Other, ABN AMRO and Other and SNS Bank and Other (see table 7).

Table 7

Bonferroni post-hoc test for banks and reputation (only significant differences)

Mdifference p Reputation ING * Rabobank ING * Other -0.50 -1.85 .035 .000 Rabobank * SNS Bank 0.84 .033 Rabobank * Other -1.36 .000

ABN AMRO * Other -1.51 .000

SNS Bank * Other -2.20 .000

(45)

These results indicate that there are differences in the levels of trust and

reputation across the banks. Customers of a bank other than ING, Rabobank, ABN

AMRO or SNS Bank, rate their bank with the highest scores on trust and reputation whereas customers of the SNS bank have the least trust in their bank and rate their bank with the lowest score on reputation.

Discussion

In today’s financial world, banks experience low levels of trust (Trust in financial services, 2014). In order to regain trust, banks have to take action. These measures should not only improve the levels of trust, but should also take the bank’s reputation into

consideration. This study had three objectives. The first objective was to provide

empirical evidence for the role of transparency in the increase in trust and reputation for Dutch banks. Second, to explore the effect of stakeholder engagement on the above mentioned empirical evidence and third, to give insight in the perceptions of customers about topics on which a bank should be more transparent. This study can be categorized in the organizational communication studies on meso level (see figure 1).

Summary of findings

Consistent with recent research by Jahansoozi (2006), Rawlins (2008), and Kim and Lee (2012), the current findings suggest a strong link between the perceived

organizational transparency amongst stakeholders and trust in the organization. Additionally, it is concluded that transparency positively affects the reputation of an

(46)

organization. The latter finding is congruent with research by van Riel and Fombrun (2007) and Rawlins (2008). While Rawlins indicated a link between transparency and the ethical reputation of a company, the evidence from this study suggests that transparency affects the complete reputation of a company.

Contrary to expectations, this study did not find a significant effect of stakeholder engagement on the relationship between transparency and trust. Nor did stakeholder engagement affect the relationship between transparency and reputation. It appeared that in the context of this study, stakeholder engagement did not strengthen the effect of transparency. Therewith contradicting indications about a link between transparency and stakeholder engagement from research by Arnstein (1969), Gray (2002) and Hooks, et al. (2002). Furthermore, the conclusion that the trust in an organization increases when transparency is accompanied by the collaboration of stakeholders (Jahansoozi, 2006) cannot be supported by the results of this study.

A plausible explanation for this unexpected result is that the qualitative findings by Jahansoozi (2006) cannot be generalised to another context. Jahansoozi derived data from interviews with members from the oil and gas industry, whilst this study was in the context of the financial industry. An alternative justification may be that collaboration of stakeholders (Jahansoozi, 2006) cannot be understood as stakeholder engagement. This however contradicts the statement that the two concepts can be perceived as the same (Greenwood, 2007). Whilst this study did not confirm the effect of stakeholder

engagement on the relation between transparency and trust, it did partially substantiate a direct connection between stakeholder engagement and both reputation and trust.

(47)

The sub-question about the topics of transparency yielded insightful results. Customers of Dutch banks indicated a need for more transparency on information that affects them directly, on information regarding their money and on information about the bank as an institution. The concerns about the topics of transparency might be caused by several incidents that took place in the Dutch financial sector and were much debated in Dutch media. According to agenda-setting theory (McCombs & Shaw, 1972; Carroll & McCombs, 2003) the salience on the media-agenda transfers to the public-agenda. This means that topics, which are widely covered in the media are also in the mind of the public (i.e. customers of banks). For example, respondents indicated a need for

information regarding compensation and bonuses of the bank’s top management, whereas this was also a much-debated topic in Dutch newspapers (e.g. NOS, 2013; NU.nl, 2014a). Even so, Dutch media covered news regarding the topic of bankruptcy (NRC.nl, 2009a), privacy (Nu.nl, 2014b), investments (NRC.nl, 2009b) and crises (Elsevier.nl, 2013a). All of these topics where also mentioned by respondents as topics that they wanted their bank to be more transparent on.

Finally, this study concluded that there are differences in the levels of trust and reputation between banks. Customers of a bank outside the four biggest Dutch banks (Bankensector-Nederland, 2014) had the most trust in their bank and rated their bank with the highest score on the reputation scale, whereas customers of the SNS bank rated their bank with the lowest scores on the trust and reputation scales. The differences between banks may again be caused by news coverage. In that case, the agenda-setting

(48)

theory (McCombs & Shaw, 1972; Carroll & McCombs, 2003) might offer an explanation for the variance in trust and reputation between banks.

In recent years the four biggest Dutch banks were all covered negatively in the news. First, in 2008 ABN AMRO almost went bankrupt but was saved by the

government (NRC.nl, 2008) five years later, the same happened to SNS Bank (NRC.nl, 2013). Later on in 2013, Rabobank endured a crisis due to the LIBOR fraud (Elsevier.nl, 2013b). Lastly, ING was covered negatively in the media due to concerns about privacy (Nu.nl, 2014b). However, other banks such as the ASN Bank (Nuzakelijk.nl, 2014) or Triodos Bank (NU.nl, 2014c) were covered less negatively or even positively in the news. ASN Bank and Triodos Bank are the fifth and sixth bank in the Netherlands so it might well be possible that respondents that are not customers of one of the four biggest banks are customers of one of these banks. Therefore it might seem plausible that people who are not customer of one of the four biggest banks rated their bank higher on the reputation scale and trust their bank more. A remark regarding this argument is that respondents could not indicate which bank they were customer of when they were not customer at one of the four biggest Dutch banks. These respondents could only choose the “other” category.

An alternative explanation for the different levels of trust and reputation between banks might be that people tend to choose their bank based on convenience (Lee & Marlowe, 2003). One of the factors underpinning convenience is the number of contact points (i.e. offices or ATM machines). Since the four biggest banks have more contact points than other, smaller, banks, it might be that customers of the other banks experience

(49)

their bank as less convenient. Despite this, they still chose to be customers of banks not in the four biggest ones.

Therefore there must have been another reason for them to switch banks. A lack of trust or a low perceived reputation may well be one of these reasons. Despite the plausibility of the above-mentioned explanations, it is important to bear in mind the possible bias in these responses due to the small sample size of the “SNS Bank” (N=9) and “other” (N=7) group.

Managerial implications

The findings in this study indicate that Dutch banks can benefit from being more transparent. The benefits include a higher level of trust from their stakeholders and a better reputation. In particular, managers of a communication department may play a crucial role in the dissemination of the relevant information to customers. It may not be necessary to include stakeholders in the decision making process of which information to send forth because this study’s results show that stakeholder engagement does not

strengthen the effects of transparency.

However, respondents demanded transparency on three broad themes:

information about issues that affect them personally, information about what the bank does with their money and information about the bank as an organization. The sub-themes, additional explanation, interest rates, risk of bankruptcy, privacy, investments, what happens with money, crisis, bonuses, employer and business conduct appeared to be of great interest to clients of Dutch banks. Since transparency increases the trust in and

(50)

the reputation of a bank, communication managers should try to be transparent on these topics in order to increase the reputation of their respective clients / employers. A communications manager should not only look for what information is disseminated by the bank, but also check whether this information reaches the right consumers.

Apart from communication managers, regulators of financial markets may also draw useful insights from the results of this study. On June 1st, 2014 the ministry of economic affairs, the regulator of the Dutch financial market, tightened its transparency benchmark (Transparantiebenchmark, 2014). This benchmark maps the transparency efforts of Dutch companies. It does so by checking its social reports, annual reports and other publicly available information. However, this research offers an alternative and possibly better (Rawlins, 2008) way to check a company’s transparency efforts, namely to ask stakeholders. This insight might be useful to regulators for improving transparency controls such as the transparency benchmark. Better transparency controls might lead to better business conduct and will therefore benefit the stakeholders of a bank

(Transparantiebenchmark, 2014).

Limitations and directions for future research

Design issues present one of the main limitations in this study. Because of the non-probability sampling technique, the validity is threatened. Moreover, since a

snowball technique was used, the educational level of the sample group turned out to be higher than the educational level of the Dutch population. Nonetheless, the remaining

(51)

statistics of the sample group showed high similarities with the Dutch population, which is positive for this research’s validity.

Another limitation that is important to note is the context of this study, the Dutch banking sector. The financial sector is a highly specific industry. This should be taken into account when results are generalised to other organizations or industries. The last limitation concerns the additional analysis of variance and is about the number of respondents for each bank. For ING, ABN AMRO and Rabobank, the number of respondents is acceptable. However, for SNS Bank and Other banks the number of respondents was low. This also explains the higher level of significance at the Bonferroni tests.

To get a better understanding of the effects of transparency, it is suggested to test the effects in other industries. Moreover, it is recommended to use the same scales that were used in this study since they were all highly reliable. However it might be wise to conduct an experiment in order to test for the causality of the effects. Additionally, the sub-themes that are derived from the sub-question in this study might offer useful material for creating stimuli in such an experiment. In this study, only two dependent variables were measured, it might be interesting to test for other effects of transparency, for example buyer intention. This study is one of the first to test the effects of

transparency from a customer’s point of view. Therefore we recommend that further research be conducted on the same line of inquiry.

Referenties

GERELATEERDE DOCUMENTEN

Op basis van het voor zover als mogelijk uitgevoerde verificatieonderzoek van de bodem in de gegraven profielputten ter plaatse van het archeologisch onderzoeksgebied

3 Value Creation and Targets Disclosure We investigate the relationship between the dis­ closure of targets and value creation. This section addresses the following: a)

In addition, this study will use the bank non-performing loans following the work of Horváth &amp; Vaško (2016) to examine the effect of the quality of institutions on

Conceptual Model Main Effects 5 Price Downgrade Consumer Judgment Quality Upgrade Extent of Consumer Orientation Depth Amount of information Limited / Extended

The primary process is controlled by the production office that is in charge of the information of an order and the chief of production who is in charge of the

At some borders in CWE there is a high volatility of the total available transmission capacity (e. border B-F). This information is relevant for market participants in order to

We identified four types of collective outcomes: a motivating goal, negotiated knowledge, pooling of resources, and trust (associated.. with positive relational experiences).

de sonde des volks geweent he[:eft] In het Bijbelboek Klaagliederen van Jeremia treurt Jeremia over de Val van Jeruzalem en de verwoesting van de tempel in 586 v. Volgens