“Whistleblowing, fraud prevention, and fraud awareness: Evidence from the Palembang Local Government of Indonesia”

Fraud is an enormous obstacle to a country’s development because it can cause failure in achieving goals. The most significant source of fraud in Palembang Local Government of Indonesia arises from the procurement of goods and services, which reached 49.1% in 2020. Therefore, this study aims to examine how fraud awareness and good government governance affect fraud prevention. This paper targets one of the Regional Government areas of Palembang City, Indonesia. One hundred twenty-two respondents comprised the head of budget staff, budget user authority staff, and commitment-making committee staff. Data were obtained using a questionnaire with a Likert scale. The study employed two endogenous variables (fraud prevention and fraud awareness), one intervening (fraud awareness), and two exogenous variables (whistleblowing and good government governance). Following this, structural equation models were used to examine the relationships between each variable. The study results show a significant positive effect of whistleblowing and good governance on fraud awareness. In contrast, fraud awareness had no significant impact on fraud prevention. The obtained results found that fraud awareness cannot mediate the whistleblowing system through fraud prevention; also, fraud awareness is not a moderating variable to influence good governance through fraud prevention. The paper contributed empirical evidence to explore fraud prevention approaches by maintaining consistency in the implementation of the whistleblowing system, anti-corruption culture, and supervision to identify unusual activities.


INTRODUCTION
Indonesia's corruption perception index ranked fifth in Southeast Asia in 2021 (Mahdi, 2022). The 2021 Corruption Eradication Commission (Dihni, 2022) report shows that South Sumatra Province has the most corruption cases. The number of corruption cases reported was 30, representing 28% of the total corruption cases in Indonesia. According to Ariyanto and Bone (2020), fraud cases often occur in Indonesian local governments. Procuring goods and services from local government agencies in Indonesia is a significant issue in eradicating corruption, given the large budget allocation of around 53% of the state revenue and expenditure budget (Ameliya, 2022).
Indonesia Corruption Watch (ICW, 2022) reported that from 2016 to 2020, there was a state loss of 5.3 trillion rupiah due to corruption in the procurement of goods and services. Judging from the number of cases, there were 1093 cases out of 2,227 cases, or 49.1%, handled by law enforcement and related to procuring goods and services. Various conflicts of interest can distort resource allocation and cause budget leakage (Rustiarini et al., 2019). Data released by the Corruption Eradication Commission show that from 2017 to 2021, the most dominant fraud cases were corruption in regency and city governments concerning the procurement of goods and services (Rustiarini et al., 2019).
Many instances of government corruption in Indonesia witness the country's poor practice of sound governance principles, especially from the control system and those non-compliant with the regulations (Gunawan, 2019). Furthermore, unhealthy practices make it difficult for stakeholders to detect fraud, and it is the responsibility of the management to implement governance and control procedures to mitigate risk (Arens et al., 2017). As Herawaty and Hernando (2020) state, good governance does not affect fraud prevention. However, there are different results, where whistleblowing does not significantly affect fraud prevention, and perceptions of justice are also needed with an open government approach (Ingrams et al., 2020). In other words, reform of the administrative structure is needed (Drápalová & Di Mascio, 2020). According to Puryati and Febriani (2020), implementing the whistleblowing system and fraud prevention, especially for state-owned enterprises in Indonesia, are relatively good categories. It is necessary to explore how the whistleblowing system is implemented in the government because the whistleblowing system can support when someone discloses information to the public or the authorities, conducts mismanagement, corruption, or other mistakes (Al-Haidar, 2018). Therefore, it is interesting to determine whether there is a relationship between the whistleblowing system, fraud prevention, and fraud awareness in each agency member recognizing the risks of committing fraud to develop a strategy for fraud prevention in Indonesia.

LITERATURE REVIEW AND HYPOTHESES
The fraud diamond theory by Wolfe and Hermanson (2004) is a theoretical framework that outlines the causes of fraud, namely pressure, opportunity, rationalization, and capability. This theory is a refinement of the fraud triangle theory proposed by Cressey (1953). When people encounter challenges, like being under intense pressure from superiors, they are likely to commit fraud (Rustiarini et al., 2019). According to Ruankaew (2016), opportunity refers to situations and conditions that allow a person to commit fraud, while rationalization is an act of justification when an individual is in an undesirable state. Capability, on the other hand, is the ability of an individual to recognize gaps as opportunities and take advantage of them. Rationalization is a risk factor that is difficult for the auditor to detect. One risk factor is a tense and uncomfortable partnership relationship between the replacement auditor or predecessor auditor and management (Rohmatin et al., 2021).
Furthermore, the fraud diamond theory provided a framework for organizations to find solutions to minimize or prevent this fraudulent behavior.  (Abazi, 2020). A law is needed to eradicate crimes such as corruption (Marwaha, 2017).
Based on the diamond theory, the whistleblowing system is related to rationalization, the underlying cause of fraud. Procurement fraud is usually triggered by the justification that the actions taken are considered correct and customary since they have become a culture in the organization (Ruankaew, 2016). As a result, an adequate whistleblowing system is required to prevent government officials from committing further fraud and effectively prevent the action generally. Referring to the National Committee on Governance Policy (Komite Nasional Kebijakan Governansi, 2008), the whistleblowing indicators are: 1. Preventing perpetrators from committing violations (principle of prevention), and 2. Promoting enthusiasm of whistleblowers (principle of early detection).
Subsequently, this study considers the application of good governmental principles from the perspective of internal government control in the procurement of goods and services, including: 1. Accountability, related to the attitude of accounting for the duties and obligations of both successes and failures that are borne by each organizational unit and individual apparatus.
2. Transparency, in terms of ensuring that the procurement process has clear procedures and each step must be followed (Rustiarini et al., 2019).
3. Responsibility, emphasizing compliance in the management of a company with the applicable laws and regulations.
4. Independence, professional management without conflicts of interest and influence/ pressure from any party that is inconsistent with the applicable laws and regulations or cannot be controlled by another person (Sunnari et al., 2021).

Fairness, righteousness and equality in fulfilling stakeholder rights based on agreements and applicable laws and regulations.
A fraud problem in an organization indicates the existence of functions that is implemented improperly. This may be because good governance is not implemented properly in the organization. According to the fraud diamond theory, fraud can occur when the perpetrator is in a situation with high pressure and opportunity, especially if the violators have low integrity. With the implementation of good governance, however, fraudulent opportunities will be curbed (Dewata et al., 2022). Following this, fraud related to procuring goods and services can also be caused by a lack of fraud awareness. This can lead to a lack of objectivity in the technical evaluation and tender qualifications. Fraud awareness plays a vital role in positively influencing the behaviors of individuals who are under pressure and have opportunities, rational reasons, and abilities. This indicates the importance of fraud awareness in curbing fraudulent behaviors. Furthermore, the awareness factor is a behavior determinant, strengthening hierarchical organizational control and reducing fraud within the organization (Zarefar & Arfan, 2017).
Even though the organization has implemented suitable fraud prevention measures, they are not implementing them due to ignorance, which can weaken awareness of fraud prevention (Abdul Latif et al., 2021). In addition, recognizing fraudulent symptoms and typology can be an effective deterrent. The emergence of fraud awareness as an investment in human capital increases sensitivity in identifying symptoms related to and discouraging participation in crime (Shonhadji & Maulidi, 2021).
Based on the literature review, the following hypotheses are proposed:

METHOD
This is quantitative research, and the primary data were obtained from questionnaires distributed to respondents. The data collection period was from May to July 2022. Using a saturated sample, the population comprises 51 members of the Regional Apparatus Organization of Palembang City. Furthermore, because three respondents were selected from each Regional Apparatus Organization based on the criterion of being a member of one of the committees for the procure-ment of goods and services, the total number of respondents amounted to 153. These committees include the budget staff, budget user authority, and commitment-making committees.
Fraud prevention (endogenous variable) consists of six indicators: strengthening the legal framework, transparent procedure, open tender/selection documents, bid evaluation, delegation of authority, and independent checks and audits (Pope, 2007 A Likert scale measurement was utilized in this analysis. The answers were assessed using five points, namely 1 strongly disagree (SD), 2 disagree (D), 3 neutral (N), 4 agree (A), and 5 strongly agree (SA). The collected data were analyzed using structural equation modeling to test the suitability of the theoretical model with the research data. Also, the significance level of each coefficient of the causal relationship was tested.

RESULTS
Out of the 153 administered questionnaires, only 122 respondents filled theirs (i.e., a response rate is 79.74%). Before testing the hypotheses, it was necessary to discover the characteristics of the respondents (Table 1).
A confirmatory factor analysis test was also conducted. Table 2 shows that the construct measurement model meets the criteria for the goodnessof-fit test. All the load factor values were generated above 0.5. Hence, the manifest is valid. Also, the values obtained for composite reliability were above 0.70. This means that all the constructs in this study were found to be reliable, and the AVE value obtained was greater than the recommended minimum value, which is 0.50. Therefore, using CR and AVE measures indicated that the construct validity and reliability were good for all variables.     The path analysis was used to determine the structural relationship and the test results through Lisrel 8.7 (Table 3). In addition, the mediating role of fraud awareness was tested, as shown in Table 3, which presents the results of two mediation models: the mediating effect of fraud awareness in the relationship between the whistleblowing system and fraud prevention and the mediating effect of fraud awareness in the relationship between good governance and fraud prevention.

DISCUSSION
Testing the first hypothesis, the results showed the path coefficient = 0.481, the t-value is 3.480, a significance level α = 5%, the value of the t-table is 0.05.119 = 1.980. Because t-count = 3.480 is more significant than t table = 1.980, H0 is rejected, or in other words, the whistleblowing system has a significant effect on fraud awareness by 0.481 so that every increase in the whistleblowing system will increase fraud awareness by 0.481 units. These findings indicate that implementing the whistleblowing system in the Regional Government of Palembang City is quite effective in assisting the disclosure of fraudulent acts reported by perpetrators consciously and in good faith, not as complaints and personal interests based on bad intentions or slander. Sopiyan (2021) showed similar findings. The practical implication of these results is that the whistleblowing system needs to be maintained, maximizing preventive actions to be more careful in all actions, especially those related to fraud. Management must also continuously assess and track relevant standard operating procedures to maintain a clean and accountable state apparatus by increasing anti-fraud behavior for everyone in the organization.
The second hypothesis demonstrates that good governance enhances fraud awareness. With a path coefficient of 0.463 between good government governance and fraud awareness, a t-count of 3.248 was calculated; the t-count (3.248) is more significant than the t-table (3.248) at a significance level of 5% (1.980). Therefore, H2 is accepted, meaning good governance influences fraud awareness by 0.463 units. Therefore, every increase in good governance will result in a 0.463 increase in fraud awareness. These results demonstrate that consistently applying ethical corporate governance principles can increase all parties' vigilance, concern, and sensitivity to reduce fraud.  2021), indicating that effective governance reduces the likelihood of fraud by anyone and the implementation of good governance will further prevent corruption. However, this result contradicts Herawaty and Hernando (2020). These findings offer practical implications for local government management to support the implementation of good governance, particularly accountability, to raise awareness of all elements within exercise control and fraud. The study results show that the third hypothesis is accepted with the whistleblowing system path coefficient on fraud prevention, t-count = 2.898 is greater than t-table = 1.980. The whistleblowing system increases fraud prevention significantly by 0.414 units for every one-unit increase. These findings indicate that the whistleblowing system, including the principles of prevention and early detection in the procurement of goods and services in the Regional Government of Palembang City, is quite effective in ensuring the delivery of critical information to the parties involved. In addition, this finding follows the rationalization of the fraud diamond theory, which states that justification triggers fraud in the procurement of goods and services in local governments. Fraudulent actions are considered appropriate and standard because they have become the organization's culture. In addition, findings show that whistleblowing is sufficient to prevent government officials from committing fraud. It means that the whistleblowing system is effective in preventing fraud in the procurement of goods and services. Adetula and Amupitan (2018) found a positive relationship between fraud prevention and whistleblowing in Nigeria. Onuora and Uzoka (2018) revealed that eradicating corruption in the public sector would be more successful with the whistleblowing policy. According to Pamungkas et al. (2017), student perceptions of the whistleblowing system significantly affect fraud prevention. Al-Haidar (2018) showed that the result of the whistleblowing situation in England is more evident than in Kuwait.
The fourth hypothesis (good governance positively impacts fraud prevention) was rejected. A t-count of 1.781 was obtained with a path coefficient of good governance on fraud prevention = 0.413. Furthermore, when the significance level was set at 5%, the t-count (1.781) was less than the t-table . In other words, good governance has no statistically significant effect on fraud prevention by 0.413; thus, every increase in good government governance increases fraud prevention by 0.413 units. According to the findings, Palembang City has not been effective in implementing good governance principles, particularly in preventing fraud in procuring goods and services. Flowerastia et al. (2021) found no significant relationship between fraud awareness, fraud prevention, and failure detection in Jakarta start-up companies. As a result, the regional governments must develop appropriate control processes, particularly independence, in supervising the procurement of goods and services to prevent and detect fraud and future prospecting costs.
Furthermore, the fifth hypothesis proves that fraud awareness does not significantly affect fraud prevention. With a path coefficient of -0.065, a t-count of -0.258 was obtained. Taking the significance level α of 5%, the t-  (2017). The theoretical implication of these findings is the need to build the awareness and sensitivity of all the parties regarding the dangers of fraud and improve control to prevent it.
The sixth hypothesis states that whistleblowing does not significantly affect fraud prevention through fraud awareness. Fraud awareness cannot fully mediate the whistleblowing system and fraud prevention. This hypothesis is based on the Sobel test results, which show that the z-count is smaller than the z-table or -0.258 < 1.98. Therefore, H6 is rejected because the whistleblowing system does not significantly prevent fraud in procuring goods and services through fraud awareness. Based on these findings, the effect of the whistleblowing system on fraud prevention through fraud awareness is 0.481 x (-0.065) = -0.031 = -3.1%. Therefore, understanding the whistleblowing system is expected to promote fraud prevention efforts. However, the presence of this system cannot be maximized if stakeholders do not have fraud awareness, primarily if the local government does not support it. These findings recommend that procurement of goods and services be reformed by emphasizing the whistleblowing system with national integrity through building synergy among all parties within the local government and by creating conditions that facilitate the detection and avoidance of fraud.
Lastly, the Sobel test results show that the z-count is smaller than the z-table or -0.258 < 1.98. Thus, H7 was rejected. The effect of good government governance on fraud prevention through fraud awareness is 0.463 x (-0.065) = -0.030 = -3%. These observations show that fraud awareness does not help good governance prevent fraud related to procuring goods and services in Palembang City Local Government, Indonesia. This is because individuals do not recognize their behavior as a member of an organization and therefore do not feel accountable for participating in organizational activities. Based on these findings, more than a concept in fraud prevention is needed. It is necessary to create fraud awareness by providing training, seminars, or workshops on the dangers of fraud, making every party in the local government possess the same understanding and knowledge about anti-corruption culture. It is also essential to pay attention to how leaders and managers understand the importance of good governance as a form of supervision to identify unusual activities.

CONCLUSION
The research objective is to determine the impact of whistleblowing systems and good government governance on fraud prevention, with fraud awareness as the intervention variable. The results show a positive and significant relationship between the whistleblowing system and fraud awareness. The effect of good government governance on fraud awareness and the whistleblowing system on fraud prevention is significant. Good government governance insignificantly affects fraud prevention and fraud awareness. The relationship between the whistleblowing system and good government governance with the fraud awareness variable as a mediating variable for fraud prevention showed insignificant results. These results indicate low awareness of not taking actions that are not commendable by enriching oneself.
The practical implications are that the oversight of the whistleblowing system is quite effective in helping fraud disclosure. Besides that, there is the awareness factor of not taking actions that are not commendable. The theoretical implications support the fraud diamond theory on the element of rationalization, which states that all forms of fraud cannot be considered correct and standard even though some people have considered it as a culture within an organization.
This study has several limitations. This analysis was conducted in Palembang City Local Government, which has its culture and work behavior. Therefore, the results cannot be generalized to other local governments in Indonesia and other countries. Future research must explore whether these findings apply to other contexts.