The Impact of Trust on the Relations between Ethical Leadership and Internal Whistleblowing Intention

The aim of this study is to examine the impact of trust on the relationship between ethical leadership and internal whistleblowing intention. The study was motivated by the lack of willingness of organization members to blow the whistle on misconduct or unethical action internally compared to external whistleblowing. The willingness to blow the whistle internally is crucial, particularly to prevent negative impacts on the organization's performance and reputation. Using a quantitative method through a survey, this study examines 140 Indonesian public servants from one of the government directorates focused on perceptions of trust, ethical leadership, and internal whistleblowing intention. Using Structural Equation Modelling (SEM) and Partial Least Square (PLS), this study unveiled that the length of working periods determines the trust development. Meanwhile, feelings are key to an affective trust that influences ethical leadership to encourage the willingness of


Introduction
The aim of this study is to examine the impacts of employees' trust on the relationship between ethical leadership and the intention of internal whistleblowing. The study was motivated by the various types of fraud found within private and public sectors even when the organisation has established a whistleblowing system. The introduction of a whistleblowing system has encouraged the revealing fraud act (Dyck et al., 2010). In practice, the numbers of unreported fraud The aim of this study is to examine the impact of trust on the relationship between ethical leadership and internal whistleblowing intention. The study was motivated by the lack of willingness of organization members to blow the whistle on misconduct or unethical action internally compared to external whistleblowing. The willingness to blow the whistle internally is crucial, particularly to prevent negative impacts on the organization's performance and reputation. Using a quantitative method through a survey, this study examines 140 Indonesian public servants from one of the government directorates focused on perceptions of trust, ethical leadership, and internal whistleblowing intention. Using Structural Equation Modelling (SEM) and Partial Least Square (PLS), this study unveiled that the length of working periods determines the trust development. Meanwhile, feelings are key to an affective trust that influences ethical leadership to encourage the willingness of organization members to intentionally whistleblowing misconduct or unethical action internally.
cases remain high. Early study reveals that the intensity of employees' trust in their leader is crucial to encourage whistleblowing (Setianto et al., 2016). Trust becomes the key to the emergence of ethical leadership because the employees believe that their leaders can stop the action. In this context, trust has had two significant effects on both ethical leadership and whistleblowing intent. Trust is an eminent factor that can increase the willingness of employees to report fraud when they believe their leaders ethically act to prevent or stop fraud. At the same time, trust is the determinant factor for the leaders presenting their ethical actions (Liu et al., 2015). When the leaders can present their ethical action, this will encourage the employees' intention to blow the whistle regarding unethical or fraud matters. From this, it is seen that trust has played a crucial role in the emergence of ethical leadership and in encouraging employees to blow the whistle. The decision to blow the whistle requires a willingness to take risks, which are determined by the person's level of trust (Gao et al., 2011;Schoorman et al., 2007). The risk is even higher when the employees are whistleblowing internally (Chang et al., 2017). The level of trust they have in the organization and its leaders is critical to their willingness to report wrongdoing. Earlier studies reveal the double-edged sword of whistleblowing as it is used to improve the organization's activities and the possibility for the whistle-blowers to deal with the potential dismissal, threats, or intimidation even when they have trust in their leaders (Brown et al., 2005;Brown & Treviño, 2006;Liu et al., 2015). Therefore, it is important to do a study to find out how trust affects the relationship between ethical leadership and employees' plans to report fraud within the company.
The negative impacts of fraud require an intensive solution. Fraud is considered a crime. Fraud occurs when a person intentionally acts or uses organisation's resources for their benefit. Fraud can be considered a general thing that occurs because of human ingenuity and is intended to gain more profit by misrepresentation (Albrecht et al., 2018). Meanwhile, others define fraud as a crime committed by an organization or person through deceit and a variety of irregularities (Rahmawati et al., 2017). From these definitions, fraud is seen as an action committed by an organization or individual that aims to gain profit through deceit through a variety of irregularities. Fraud is found in both the public and private sectors. Fraud also occurs in most levels and layers of classes and employees. While fraud can be committed by a person intentionally, it commonly occurs in collaboration with other parties. Bribery, misuse of assets, and manipulation of tax, payment, or financial reports are considered fraud committed by parties or organisations. These have caused some significant impacts on the organization. These negative impacts have driven the establishment of whistleblowing systems in a number of organisations. The main purpose of this whistleblowing system is to encourage the organization's members to report illegal, violation, or immoral activities (Khan, 2009).
Despite the implementation of a whistleblowing system, the organisation must increase the level of employee trust. Trust is the positive conduct exhibited by employees in an organization. A high level of trust within an organisation leads to decline disputes amongst employees, absenteeism, stress (Farahbod et al., 2013). Trust in organisation develops from individual belief as a result from their relations and experiences within the organisation. This positive feeling also emerges from the social emotional exchange which leads to the acceptance and feeling of secure which strengthen the relations (Dyck et al., 2010;Zhu et al., 2013). This is why trust is often considered as a tool to improve the employee motivation and innovation work environment as well as to present more ethical and accountable actions (Farahbod et al., 2013). Therefore, the establishment of whistleblowing system encourages the organisation members to be more aware to open up the unethical behaviour such as fraud in the organisation. Early study reveals there are two main factors that can prevent fraud; honesty culture and minimising loopholes for committing fraud (Albrecht et al., 2018). Honesty culture as presented through trust is the key in whistleblowing system particularly trust that is developed amongst the organisation members.
The number of participants who report unethical behaviour is still small, despite the growing trust. Early research shows that dismissal and retaliation discourage employees from reporting wrongdoing or unethical behaviour. (Chang et al., 2017;Latan et al., 2018;Miceli & Near, 2013;Zhang et al., 2009). There are many factors that prevent the organization's members from reporting the unethical behaviour. One of the reasons is their lack of trust in and obedience to their leaders. The level of employees' trust in their leader determines the employee's confidence to express their opinion regarding fraud. The level of an employee's trust in their leader is also a reference for the employee's willingness to blow the whistle. The higher level of their trust drives their belief that the leader will do or conduct the necessary action to prevent or stop fraud. In practice, a number of employees may not be able to express their opinion regarding fraud. The risks and impacts of their actions become the factors that prevent them.
The employee may assess the risk and negative impacts of their actions and their relationship with the leader. Retaliation or discrimination is the most common impact of their whistleblowing action. These are mostly driven by the employees to remain silent or not to take any action when the risk negatively affects them. (Chang et al., 2017;Liyanarachchi & Adler, 2011). Whistleblowing may not work as expected when the employee feels threatened by these negative impacts. From this, it is seen that trust becomes a crucial factor in motivating the employees' willingness to take whistleblowing action (Liu et al., 2015). It is crucial for both private and public organizations to understand the role of members' trust in the leaders to reduce and prevent fraud. Referring to this argument, it is evident that a leader's followers' trust is essential for successful whistleblowing, even though ethical behaviour does not guarantee followers' trust and readiness to report unethical behaviour. This situation leaves a gap of knowledge in the extent to which the trust on the ethical leader affects the organisation members to present their whistleblowing intention. This situation creates a knowledge gap regarding the extent to which trust in the ethical leader affects the ability of organization members to present their whistleblowing intentions. This study contributes to the accounting, governance, and business ethics areas. The following sections present the literature study and the hypothesis development; the research method; the results and discussion; and the conclusion.

Literature review and hypothesis development
To analyse the impact of ethical leadership on employees' willingness to do whistleblowing, it is crucial to understand the framework of social learning theory. The theory is selected to provide a framework to understand the relationships, changes, and behaviour presented by a leader and their followers. The theory puts a lot of emphasis on how the way employees or followers are treated in an organization can change their minds. (Bandura, 1977). The treatment received by the whistleblowers becomes the main encouraging factor for change. This treatment can be done through a learning process that includes modelling and experience. In the case of an organization, the members usually observe and emulate their leaders' behaviours when they see certain behaviours modelled (Bandura, 1978). This process of learning happens when the followers directly observe their leader's behaviour through their performance or modelling. The characteristics are observed by their followers become the new behaviour patterns or guided for action as well as their personal perception toward their leaders. Akers and Jennings (2016) emphasis imitation is one of learning processes when individual observation toward others behaviour. This observation becomes an influence factor on the employers or members' decision when deals with the wrongdoing or unethical behaviour. Besides the observation, the members' interaction and association with peers is also influence their decision when it comes to participating or conforming certain behaviour, action, or situation (Akers & Jennings, 2016). Interaction occurs as a result of their time spent together. Through this interaction, values, attitudes, and orientations develop, which affect the member's perception toward others. This social learning theory is used as a framework to figure out what it means for a leader to be ethical and what motivates employees or followers to report wrongdoing.

Ethical leadership and whistleblowing
Ethical leadership becomes the key to improving trust and the intention of organisation members to engage in whistleblowing. Leadership as an organisational function aims to influence and motivate the members to chase the organization's goals (Pride et al., 2019). Leadership roles are critical in developing and implementing an ethical culture, as well as monitoring its effects within the organisation. Consequently, a leader has sufficient control and power, which also becomes a challenge for them to justify and use the power (Becker, 2019). The ability to balance control and power determines the characteristics of a leader. Responsibilities, trustworthiness, credibility, care, and fairness are the main characteristics of ethical leadership (Becker, 2019). None of the leaders can present these characteristics perfectly.
Therefore, ethical leadership is mainly based on the ability of the leader to have a positive impact on the organization and its followers.
Failure to present positive impacts on the organisation and its followers causes the emergence of the negative side of power. Power abuse and corruption are examples of the negative side of power that may occur when the leader fails to present positive impacts. Early studies reveal that ethical leadership mainly engages with being trustworthy and being honest. (Brown & Treviño, 2006). Trustworthiness and honesty are based on the explicit actions of the leader. This clear action is interpreted as a message to the followers to draw their attention to how moral leaders communicate their message. The action is seen by the followers, who become the models. A study on social science focuses on how people can learn by watching and copying what others do, and how this becomes the basis for moral guidance (Brown & Treviño, 2006). In addition to action, communication is included as a factor that is used to influence the followers. Early studies reveal that the main characteristics of ethical leadership are action and communication. (Lawton & Paez, 2014). Action is concerned with behaviours and personal conduct, whereas communication is concerned with relationships. This means when the ethical standards are set, the leader communicates them to the followers (Brown & Treviño, 2006).
Focusing exclusively on the characteristics of ethical leaders, honesty is the most common factors to discuss amongst scholars. Honest leads to trust which becomes a key for the emergence of ethical leadership. Trust in leader is a factor in leadership particularly when deal with salient problems. Loyalty and commitment from followers appear when they believe they can trust their leader. The ability to influence and become a role model determines the successful of leaders to mobilise others to achieve the organisation's goal.
The requirement to achieve the organisation's goal requires normative behaviour through action and interpersonal relationship presented by the leaders. This is in particularly to reduce the unethical behaviour as well as enhances organisational citizenship behaviours (Resick et al., 2006;Walumbwa & Schaubroeck, 2009) which mainly related to developing moral and virtuous behaviours of others. Failure to present these ethical actions leads to organizational collapse and some social consequences. (Ciulla, 2014). The ability of a leader to present and direct their followers is seen as genuinely caring about their well-being. These actions affect the follower's decision and perception, particularly when they will present their intention to do whistleblowing about wrongdoing, because their intention is influenced by their leader's value, perception, and opinion.
Despite their belief and trust in their leaders, organization members' willingness to blow the whistle on unethical behaviour is also hampered by their uncertainty about the consequences of their actions. Employees are afraid of being fired, threatened, intimidated, or losing their jobs if they tell their leader about unethical behaviour within the company. (Chang et al., 2017;Liyanarachchi & Adler, 2011). The ethical roles and support from leaders are needed in this situation, particularly to reduce the employees' fear (Walumbwa & Schaubroeck, 2009) to report the wrongdoing internally. The extent to which the leader's ethical roles and support affect the employees' willingness to internally blow their whistle needs further evaluation. The fear feeling caused by the employees' uncertainty and lack of trust in their leaders prevents the success of internal whistleblowing. Referring to this situation, the hypothesis is developed as: H1: Ethical leadership improves employee's whistleblowing intention internally

The mediating effect of trust in the relation between ethical leadership and whistleblowing
Trust is often seen as a commitment between two or more people or parties. Trust develops, mostly, from credibility, integrity, character, and the ability to develop interpersonally amongst parties involved. Early research defines trust as a willingness to be liable to others based on positive expectations about the intentions and/or behaviour of others (Mishra et al., 2019). From this definition, trust relies on the willingness, openness, and positive expectations of both parties involved. Meanwhile, a relationship is a complex structure. The level of relationships determines the level and nature of trust amongst parties. As a result, the challenge appears when each person may have different expectations and willingness. Diversity is the main factor that contributes to these differences. Failing to resolve the differences does not only reduce the trust amongst parties, but it also endangers and risks the future of the organisation.
Trust is crucial within organizations as it is needed amongst the organization's members based on their interpersonal relations. Prior studies show that trust improves employees' willingness to do beyond their job expectations (Berkovich, 2018;Newman et al., 2014). When employees trust their leaders, they are more willing to commit to their organization (Chiang & Lin, 2016) and perform better (Newman et al., 2016;Saleem et al., 2020). As a result, employees' motivation to take ethical actions is also improved. (Pulungan et al., 2021). This type of trust also requires sufficient support and effort from the leader. The ability of a leader to build trust in his or her followers is said to reduce uncertainty and positive outcomes, as well as to facilitate collaboration and innovation and to improve performance outcomes. (Mishra, et al., 2019). Trust contributes significantly to the performance of an organisation. Early studies show that trust is a key for organizational performance, which determines the willingness of voluntary cooperation when control and command do not work properly (Bijlsma & Koopman, 2003). This definition emphasises that trust can be a tool to prevent a person from fraud or unethical action when the organisation's control is poor. However, this crucial trust often does not meet the expectations. Some factors contribute to the development of trust within the organization. The importance of emotions has developed trust into two forms: affective and cognitive trust.
Affective trust focuses on an individual's feelings toward others. Affective trust is defined as the emotional feeling toward others based on rational judgment where emotional concern and attention are required (Gobena & Van Dijke, 2016). Affective trust involves not only person's emotion, it also involves rational and logic judgment to develop and present a feeling toward others (Pulungan et al., 2021). Consequently, affective trust is also seen as the liableness (Kim, 2005) toward others which become the reference for social interactions. Aside from liableness and emotional feelings, affective trust is frequently viewed as a reliance in which feelings of security and perception serve as the relationship's reference point. (Johnson & Grayson, 2005). Affective trust from this perspective is more about emotional reliance on others. Therefore, emotional feelings and attachment toward others are key factors in examining the extent of affective trust.
Cognitive trust is another emotional feeling related to the other reliance. The reliance on others develops from their experiences and knowledge which enable the person to confidently predict the action and obligation (Fulmer & Dirks, 2018). Cognitive trust is another emotional feeling related to the other reliance. The reliance on others develops from their experiences and knowledge, which enables the person to confidently predict the action and obligation (Johnson & Grayson, 2005). Focusing exclusively on the initial interactions, knowledge is the key factor in establishing cognitive trust. Knowledge plays a significant impact on measuring the level of trust, which is also a reference for future actions. A judgment on certain actions is measured based on the response, observation, or report received from the relationship. The strong response or report received determines the perception and trust.
Whistleblowing globally develops following the increase in accounting violation cases. The willingness of organization members to blow the whistle becomes a key intention since whistleblowing has successfully prevented fraud and accounting violation cases. Whistleblowing is defined as the disclosure by organization members of illegal, immoral, or illegitimate practices under the control of a person in an organization which may affect the action of (Owusu et al., 2020;Valentine & Godkin, 2019). Meanwhile, there are many reasons that driving the emergence of whistleblowing; incentive (Dyck et al., 2010), moral obligation (Valentine & Godkin, 2019), or individual behaviours (Owusu et al., 2020). While a number of whistleblowing incidents occur through external access, internal access is the most common access taken by organization members. Early research indicates that the attitude influences the intention to blow the whistle, which becomes an important predictor of ethical intention (Owusu et al., 2020). However, this attitude is not fully the key factor for an individual's intention to blow the whistle. The value and opinion of their leader also contribute to their decision to blow the whistle. In this situation, their trust towards the leader becomes crucial.
Strengthening relationships between leaders and followers leads to an increase in trust. Security, care, and concern demonstrated by the leader may influence the emotional feelings of followers (Johnson & Grayson, 2005). As the emotional connection develops and deepens toward the leader, the perception of trust motivated the development of affective trust. The level of trust is not only influenced by the emotional feeling. The competence and reliability of leader also determine their follower's trust. Competency and reliability become a knowledge driven for cognitive trust. These competency and reliability are the foundation for leader reputation and the emergence of the followers' knowledge to judge their willingness and reliance. While the ethical leader is considered to have significant influence on the emergence of intention to blow the whistle, there is limited information in the extent to which trust affects the emergence of intention to blow the whistle to their ethic leader. Moreover, there is limited information and research regarding the influence of affective trust and cognitive trust on the ethical leadership and the intention of organisation member whistleblowing internally; whether the affective trust is stronger than cognitive trust or vice versa. Based on this gap of knowledge, the hypotheses for this study are developed as: H2: Affective trust mediates the effect of ethical leadership on whistleblowing intention internally. H3: Cognitive trust mediates the effect of ethical leadership on whistleblowing intention internally.

Research method
This study applied a quantitative method and used questionnaires to collect data from the respondents.
The respondents in this research worked in a regional department of finance which had an established whistleblowing system. All the respondents must have had at least three years of working experience because employees require a certain period to be familiar with and establish certain relationships with their leaders. Due to restricted data access, the researcher was unable to locate population numbers matching the criterion. Therefore, convenience sampling is used in this study. These approaches were developed because it was effective to scope non-probability sampling selection when no population information is available (Hair, Jr. et al., 2020).
This study applied a quantitative method and used questionnaires to collect data from the respondents. Using quantitative method enables to reach more respondents to have to their opinions regarding trust, ethical leadership, and whistleblowing intention. Respondents are civil servants in Indonesia who have worked for at least three years as public servants with a minimum age of employees is 25 years. The reasons why the participants must have at least three years of working experience is to ensure that the respondent has sufficient knowledge about their leader and working environment. While fraud is found in both public and private sectors, the selection of public sector and civil servants as the focus of this study is due to high pressure from the public on the fraud issues and the effectiveness of the whistleblowing system within government organisations.
Unfortunately, since participants in this research must work at least for three years in their current organization, there is publicly available information about the population that meets the criteria. The researchers, then, select the convenience sampling method. This method is specifically used for nonprobability sampling selection in the absence of population characteristics (Hair, Jr et al., 2020). With the convenience sampling technique, (Sugiyono, 2010) suggests that the approximate benchmarking sampling is between 50-300 samples. This study collects 140 respondents.
A survey is employed through online questionnaires to civil servants who work in one of the governments' directorates between April to May 2020. The online questionnaires were sent due to the Covid-19 pandemic situation. One hundred seventyseven (177) respondents returned their questionnaires. However, only 140 questionnaires could be used for further analysis. The other 37 respondents could not be analysed mainly because they had working experience less than three years or the questionnaires were incomplete.

5-point Likert Scale
Based on the literature review, the questioners focus on ethical leadership, affective trust, and cognitive trust as the independent variables, while internal whistleblowing intention is the dependent variable. The questioners are developed using Likert Scale of 5 from strongly disagree (1) to strongly agree (5). The questioners are divided into four sections based on the variables.
These questioners are adapted from previous studies for each variable as shown on Table 1. Table  1 summarizes the measurement instrument of variables. The data are analysed to examine the validity. For this result, statistic tools are employed to examine these results. These results are used to examine the hypothesis using the Structural Equation Modelling (SEM) and the Partial Least Square (PLS) tools.

Results and discussion Respondents' profiles
To ensure that the participants recognize their leaders well at their current company, they must indicate their working period at the company. Table  2 shows that more than 60% of the participants have worked at the company for at least 11 years, while 31.4% had worked for around 6-10 years. Only 7.9% of the respondents worked at the company for 3-5 years.Apart from working experience, the survey also collected other demographic information including gender, age, and education level. Most participants in our survey are male (86.4%) The majority of respondents are also between 40 to 50 years old (45%). Only very few participants who are more than 50 years old. The participants hold at least diploma degree. Most of them (88.3%) had earned at least an undergraduate degree.

Validity and reliability test of the measurement model
The study applied SMART PLS 3 for the data analysis because it is a non-parametric statistical tool used to analyse a relatively small sample size and complex models without distributional data assumption (Hair, Jr et al., 2020). The study runs several tests to examine the inferential statistics. The convergent validity as shown on Table 2 demonstrates that most variables have external loading value bigger than 0.7 except for EL8 variable. The EL8 presents its 0.661 which is below 0.7. As a result, this EL 8 is excluded from the further test. In addition to convergent validity test, this study conducts construct validity test using the average variance extracted (AVE). The result shows on Table 3 where all variables have bigger than 0.50 points. This means that all variables are valid since they have more than 0.50 points. Discriminant validity is used as to test whether there is latent variable to other variables based on the correlation amongst these latent variables. A variable is considered to have discriminant validity if its correlation indicator is bigger than correlation value of the other variables. The measurement is identified through the cross-loading values among the variables. An indicator is considered to meet the discriminant validity when the variable's crossloading value is bigger than other variable. From Table 4, it can be seen that each variable has bigger cost-loading value toward other variables which indicates that each variable has sufficient and good discriminant validity in comparing with its respective variables.

Hypothesis test results
The inner model test is done to present the construct relations amongst variables prior the hypothesis test is conducted. For this study, trust is divided by affective trust and cognitive trust which are tested to measure their intensity and influence on the relationship. The study consists of five variables. The independent variables are affective trust, cognitive trust, and ethical leadership, while dependent variable is the internal whistleblowing intention.
The determinant coefficient test aims to test the extension impacts of independent variable on dependent variables as shown on Figure 1. From Figure 1, the inner model test demonstrates R Square results which relatively moderate. The ethical leadership may explain 56.1 percent of the variance in affective trust, while the rest is influenced by other variables. Meanwhile, ethical leadership may explain 63.1 percent of variance in cognitive trust, while the rest is influenced by other variables. For whistleblowing factor, the ethical leadership, affective trust, and cognitive trust may explain their 34 percent of variance on internal whistleblowing intention, while the rest is affected by other factors.
The hypothesis test is done through bootstrap model which aims to minimise the abnormalities problem found from research data as shown on Figure 2, and Tables 6 and 7. The bootstrapping results show that ethical leadership does not directly affect employees' intention to report fraud internally (EL → IW; β = -0.051; p = 0.784), which means that the first hypotheses is rejected. Ethical leadership could only improve employees' internal whistleblowing intention indirectly, particularly when the relationship between both variables is mediated by affective trust.
Ethical leadership has significant and positive effect on an employee's affective trust (EL → AT; β = 0.749; p = 0.000), and then affective trust significantly and positively affects employees' intention to conduct whistleblowing via internal channels (AT → IW; β = 0.542; p = 0.000). This indicates that the second hypothesis is accepted. On the other hand, cognitive trust does not mediate the effect of ethical leadership to employees' internal whistleblowing intention since the effect of cognitive trust to internal whistleblowing is insignificant (CT → IW; β = 0.106; p = 0.337). Thus, the third hypothesis is rejected. Based on these hypothesis results, the analysis of mediating effects is continued. As shown on Tables 6 and 7, the path analysis reveals that affective trust is able to mediate the relations between ethical leadership and internal whistleblowing intention as the ethical leadership causes the increasing of internal whistleblowing intention when the affective trust is existed as shown on Tables 6 and 7. Tables 6 and 7 show that ethical leadership may not have negative or no impacts on the internal whistleblowing unless the affective trust mediating their relationship. In contrast, cognitive trust does not seem to have any influences on the relationship between ethical leadership and internal whistleblowing intention. This is occurred as the cognitive trust presents insignificant impacts from this test.

Discussion
This study emphasises several important results about internal whistleblowing intention amongst public servant officers. The study results reveal that ethical leadership cannot directly influence the employees' intention to do internal whistleblowing. The impacts on employees are more indirect effects through affective trust.
There are several reasons why the ethical leader fails to affect the emergence of whistleblowing directly. First, the direct supervisors who are evaluated by the respondents through questionnaires may not be the top management of the organization where the respondents work. In other words, these direct supervisors may be subordinate to higher officers within the organization. This can be a problematic issue when the leaders do not show ethical characteristics, especially in organizations where fraud has occurred systematically, including in government institutions (Prabowo & Cooper, 2016). Since Indonesian culture is likely to focus on collectivism and high power distance culture, (Insight, 2020), the employees are even more incline to report the whistleblowing when their top leaders are corrupted. They keep themselves to be part of the organisation or group.
Secondly, retaliation threats can prevent the employees from doing internal whistleblowing, despite their direct supervisors' ethical characteristics. Retaliation can be in the form of transferring to other places or threats. Retaliation threats are still found in the form of social pressure, downgrading, or losing their job, even though countries have developed legal protection for whistle blowers. (Chang et al., 2017;Liyanarachchi & Adler, 2011).
Since these retaliation may have some consequences, the employees may evaluate its positive and negative outcomes (Cassematis & Wortley, 2013). If the outcome has more negative impacts, the employees are likely to remain silent. In the case of employees who perceive that their top leaders are unethical, they may choose not to do internal whistleblowing even though their ethical direct supervisor motivates them to report fraud as early as possible. Early study reveals that the power and dependence of whistle-blowers against the wrongdoing can be a factor that the employees to remain silent (Miceli & Near, 2013).
Thirdly, in the fraud systematically environment, any internal reports about fraud may not be followed up. Subordinates tend to report fraud by using the internal channel when they perceive the internal parties process their reports fairly (Anugerah et al., 2019;Liu et al., 2015). When the top leaders' wrongdoing is reported subjectively and unjustly, the employees may not report the fraud internally. Our findings show that respondents' direct leaders with ethical characteristics may improve the respondents' willingness to do whistleblowing when they have affective trust toward their leaders. In other words, affective trust mediates the relationship between ethical leadership and whistleblowing intention. Affective trust develops from the time spent and interactions in the organization (Akers & Jennings, 2016). Through this interaction, the employees observe their leaders and other members. According to social learning theory, employees will observe and emulate their leaders' behaviours when they see certain behaviour as modelled. (Bandura, 1978). With affective trust, the employees imitate their leaders' behaviours not only due to the leaders' role models. It is also because they have the emotional attachment with their leaders. Through this observation, the employees also learn consequences of these performance and model (Akers & Jennings, 2016) which may encourage them to engage and develop trust on their leaders. If the outcome has more negative consequences, it is likely that the employees will remain silent. Employees who perceive their top leaders to be unethical may choose not to engage in internal whistleblowing, even though their ethical direct supervisor encourages them to report fraud as soon as possible.
The relations between employees and the leaders are a key for their trust relations which determine the employees' perception to reduce the retaliation's effect (Rehg et al., 2008). Employees may feel less concerned about retaliation outcomes if leaders can create a safe environment in their workplace. This safety environment will encourage the employees to do whistleblowing. (Liu et al., 2015).
In contrast to affective trust, cognitive trust has less influence to drive the employees to do whistleblowing. This study consistent with Lu (2014) which notes there is no influence of cognitive theory on organisational citizenship behaviour. This is because the cognitive trust is likely to emphasise rationality and logic. However, Indonesian culture is strongly influenced by the principles of indebtedness or debt of kindness, which focus on mutual obligation (Prabowo & Cooper, 2016).
Referring to Tables 6 and 7, the ethical leadership presents negative or no impacts on whistleblowing intention. This result differs from the existing ethical leadership and whistleblowing intention theory. The leaders should motivate their employees behaviours particularly through the emotions such as leader's trust (Yuan et al., 2021).
While the study enables to reveal the impacts of ethical leadership on the emergence of both affective and cognitive trust, it fails to show the impacts of ethical leadership on internal whistleblowing intention. This is supported by the fact that ethical leadership has only a 34% influence on internal whistleblowing intention. This means that ethical leadership may not be the only thing that makes employees want to blow the whistle.
The study also reveals the influence of emotional feeling. Early study reveals the influence of emotional feeling, and competence and reliability play significant effects on the development of trust (Johnson & Grayson, 2005). Affective trust is seen to have more impacts on the development of trust. The roles of emotional connection is stronger particularly when followers have strong feeling about security, care and concern which are presented by their leaders (Johnson & Grayson, 2005). Emotional feeling which develops through the length of the working period has emerged in the followers' trust that makes the leaders easy to convince their followers to blow the whistle the unethical action (Al-Olimat, 2020;Alpkan et al., 2020;Anugerah et al., 2019).
In summary, ethical leadership's effects on whistleblowing are not just influenced by their actions. In practice, this activity does not enable ethical leaders to persuade their followers to report unethical behaviour. The majority of a group's trust in its leaders is based on their emotional connection with them over a long period of working together. Trust must be established between leaders and followers. To build rapport, leaders should gradually increase their interaction with their followers.

Conclusions
This study aims to examine the impacts of trust on ethical leadership and internal whistleblowing intentions. Our study shows two significant findings. First, ethical leadership alone does not affect the employees' intention to report any fraud. This may be caused when the leaders are not the top leaders of the government's organization. Second, our study shows partial impacts of trust on ethical leadership impact on employees' internal whistleblowing intentions. Only affective trust can improve the effect of ethical leadership on employees' whistleblowing intentions. When employees have emotional faith in their leaders, they are more likely to report fraud internally.
This finding adds to the empirical evidence that affective trust helps leaders present their ethical actions to their employees. In practice, trust is developed from credibility, integrity, character, and the ability to develop interpersonal relations among parties. The leaders could thus promote internal whistleblowing intentions amongst employees once affective trust is established amongst them.
This study contributes to the accounting literature by extending prior studies about fraud and whistleblowing. The results show the importance of affective trust in mediating the effect of ethical leadership on internal whistleblowing intention. Leaders can promote internal whistleblowing among their employees if their employees establish an emotional bond with and trust in their leaders. This finding should also be useful for organizations in the governmental sector. Having an ethical leader is necessary to improve whistleblowing. However, developing trust between leaders and their employees cannot be neglected. The higher the emotional trust held between leaders and subordinates, the more accurate the exchange of information, the better understanding of performance goals, and the quality of communication, which finally may increase internal whistleblowing intentions of employees.
This research is not without limitations. First, the sample is relatively small. This creates opportunities for future studies to use more samples from more government organizations to gain more insight into the role of trust in promoting whistleblowing. Second, future studies may analyse real whistleblowing cases to investigate why a whistle-blower decided to report fraud. The studies can be done quantitatively and qualitatively to gain a better understanding of a whistle-blower's motivation.