“The problem of corruption in government organizations: Empirical evidence from Indonesia”

Corruption in government organizations is an important and relevant topic to study because of its impact on the state in terms of financial losses and a decrease in the quality of human development. This study is also relevant because previous analyses are still limited in their modeling and measuring comprehensive fiscal decentralization variables. This study aims to examine the effect of fiscal decentralization and quality of government on the level of corruption and the impact of corruption on the human development index. The sample of this paper comprises 113 local governments on the island of Java, Indonesia, for the period 2015–2018. Statistical testing was carried out using Partial Least Squares-Structural Equation Modeling (PLS-SEM). The test results show that fiscal decentralization has a positive effect on corruption with a path coefficient of 0.100 and is significant at 5% alpha. Likewise, poor governance (proxied by internal control weaknesses) has a positive effect on the level of corruption with a coefficient of 0.062 and is significant at an alpha of 10%. The results of the PLS-SEM test also show that corruption has a negative impact on the human development index with a coefficient of –0.206 and is significant at 1% alpha. The practical significance of this study is the importance of the internal control system reliability as a complementary variable for fiscal decentralization to prevent corruption.


INTRODUCTION
The problem of corruption in government organizations nowadays is widely recognized as a major ailment that threatens not only economic development but also the foundations of society. It is believed that a high level of corruption in government organizations, in the long run, will reduce the welfare of people in general, not only affecting the weakest classes of society through inequality and poverty but also access to various public services such as health services and education services. Various government corruption scandals occurred in various countries including Spain, France, Guatemala, Honduras, and so on, which attracted the attention of public sector researchers and practitioners (Cuadrado-Ballesteros et al., 2020). The problem of corruption in government organizations also occurs in Asian countries, including Indonesia. The level of corruption in local government organizations in Indonesia is relatively high. According to data from the Corruption Eradication Commission (KPK), from 2014 to 2019 there were 97 criminal acts of corruption carried out by regional heads in local governments in Indonesia. The problem of corruption has caused the state significant losses. Indonesian Corruption Watch (ICW) reports that the value of state financial losses in 2019 reached IDR 8.4 trillion, an increase from 2018 with state financial losses of IDR 5.6 trillion. The seriousness of the problem of corruption in Indonesia is also shown based on the Corruption Perception Index (CPI) released in 2020 by Transparency International (TI).
One of the factors suspected of causing high levels of corruption in local government organizations is fiscal decentralization. Since the reform era, the administration of local government in Indonesia has undergone significant changes, namely a shift from centralization to decentralization. With the implementation of fiscal decentralization, local governments must be able to play their role in managing their finances independently so that all potential can be maximized through effective and efficient planning. However, in fact, the implementation of decentralization raises many problems, especially in the material use of the regional budget, which can be indicated by criminal acts of corruption such as the regional expansion budget, disaster management, and working visits. The modes carried out include, among others, the misuse of the regional budget, inflating funds in the procurement of goods and services, and making fictitious projects (Nirwanto, 2013).
The effect of fiscal decentralization on corruption has become a topic of interest for public sector research in various countries. However, the empirical evidence of several previous studies related to the effect of fiscal decentralization on corruption is still not consistent. This inconsistency requires further evaluation in the conceptualization and measurement of fiscal decentralization, which is not yet comprehensive. Fiscal decentralization is an unobserved construct consisting of several elements or dimensions to capture aspects of local government revenues and expenditures. With the authority and flexibility of local governments in managing or regulating regional finances provided by the central government, it is necessary to follow up with strong supervision and control to avoid irregularities and fraud. One of the ways to strengthen the supervisory function is through the establishment of an adequate internal control system as a prerequisite for the governance of government organizations (Mardiasmo, 2018). Therefore, research modeling that examines the effect of fiscal decentralization on corruption should also include the variable of governance quality such as internal control. In previous research, fiscal decentralization and internal control systems as determinants of corruption were analyzed separately. The separation of these two variables can result in the misspecification of the model because poor quality governance can hinder monitoring incentives, so fiscal decentralization can actually encourage an increase in corruption. In this case, there is a decrease in the benefits of fiscal decentralization to prevent corruption. This study argues that fiscal decentralization and quality of governance are complementary variables as determinants of the phenomenon of corruption, so both should be included in the model. This study is relevant because it is an extension of previous research in several respects. First, in its modeling, this study uses the variables of fiscal decentralization and quality of governance (proxied by internal control) as complementary determinants of corruption in local government organizations. Second, aside from the modeling, this paper also develops previous research in a more comprehensive conceptualization and measurement of fiscal decentralization. With reference to Hair et al. (2017), fiscal decentralization is conceptualized as an unobserved construct as measured by several formative indicators. Second, this study includes the variable of governance quality, which is proxied by the reliability of the internal control system. Third, the measurement of the level of corruption in this study is different from several previous studies that mostly used the corruption perception index. The level of corruption in this study is measured by using the real level of corruption that has been decided by the Supreme Court so that it can better describe the actual phenomenon of corruption cases. Fourth, this study also examines the impact of corruption on the human development index of local governments in Indonesia. This topic is still limited to researching the Indonesian context because previous research has mostly focused on the determinants of corruption. The effect of corruption on human development is important to study because of its impact on all these aspects. Akcay (2006) states that corruption hinders all aspects of human development.

LITERATURE REVIEW
According to agency theory, the main characteristic of the relationship between principal and agent is the existence of a delegation contract. Lupia and McCubbins (1994) state that such a contract emerges when one or several persons (or principal) choose another person/persons to work following the needs and wishes of the principal. Zimmerman (1977) states that all organizations, including public and private sector organizations, can experience agency problems. This theory assumes that there is a lot of information asymmetry between the agent (local government) who has direct access to information and the principal (the community). This information asymmetry that occurs allows fraud or corruption by agents. To avoid agency problems, accountability and supervision are needed. In agency relationships, it is undeniable that agency problems will occur. Agency problems that can occur in the public sector are information asymmetry related to the use of regional finances, which leads to corruption by agents (local governments) that harm the principals (the communities). According to the fraud triangle theory, there are three causes of corruption, one of which is an opportunity that occurs because of the abuse of authority that has been used. Decentralization is one of the opportunities that can lead to corruption. Decentralization is defined as the transfer of authority and duties from the central government to local governments to administer and regulate the government. Fiscal decentralization as a result of the decentralization policy plays an important role in carrying out the authority and functions of local government. The delegation given by the central government to local governments is expected to be able to explore and utilize all the potentials of the regions so that they can be used for the benefit of the wider community. However, in reality, the implementation of fiscal decentralization in Indonesia seems to transfer the incidence of corruption from the central government to local governments. The number of regional heads involved in corruption cases is increasing as is the number of corruption cases that occur in local governments.
In addition to agency theory and fraud triangle theory, the phenomenon of corruption in government organizations can also be explained by public choice theory. Buchanan (2009) states that public choice theory can offer useful insights because it explains the behavior of interacting agents in political markets. According to this theory, politicians can be selfish, rational, utility maximizers and try to find ways to maximize their own welfare, rather than the public interest. In order to overcome the opportunistic actions of politicians such as corruption, it is necessary to adopt institutional arrangements. New public management techniques such as good governance are institutional arrangements to minimize political opportunism, including corruption and rent-seeking (Mbaku, 2008).
The objective of decentralization is the delegation of financial affairs from the central government to local governments with the aim of, among other things, providing more adequate monitoring by the local public (Faguet, 2014;World Bank, 2001). With closer monitoring of the electorate, decentralization is also an important anti-corruption NPM mechanism, and most international organizations, such as the World Bank and the International Federation of Accountants, continue to support fiscal decentralization projects. Decentralization is considered an anti-corruption mechanism because, in smaller governments, voter monitoring is likely to be more effective. After all, local voters have more complete information to monitor and discipline local politicians (Fan et al., 2009). On the other hand, there is an argument that fiscal decentralization can increase the possibility of corruption. This is because local politicians may face more pressing demands from local political elite interest groups than national bureaucrats face and therefore tend to act corruptly ( (Fan et al., 2009). To further clarify the mixed findings in the decentralization literature, public sector management research needs to include other determinants of corruption in the research model (Changwony & Paterson, 2019). This study includes a determinant of corruption in addition to fiscal decentralization, namely the quality of governance, which is proxied by internal control. With the phenomenon of the strong dominance of local political elites, this study argues that in the context of Indonesia, fiscal decentralization can encourage corruption. Fiscal decentralization is an opportunity factor that drives corruption in local government organizations in Indonesia. Several previous studies provide empirical evidence that the higher the fiscal decentralization, the greater the level of corruption. Shon and Cho (2020) examined the relationship between fiscal decentralization and corruption in state and local governments in the United States. The results provide evidence that fiscal decentralization will increase the occurrence of corruption: that is to say, there is a positive relationship between fiscal decentralization and corruption. Research in Indonesia also shows that fiscal decentralization has a positive effect on the level of corruption.  2021) showed that regional independence in the financial sector has a positive effect on corrup-tion. Therefore, supervision of the implementation of fiscal decentralization must be carried out adequately to prevent corruption. The results of this study support the findings of Astuti and Adrison (2019), and Yanto and Adrison (2020) that the large budget managed by local governments can lead to an increase in corrupt behavior.
According to the Corruption Discretionary Monopoly Accountability (CDMA) theory developed by Klitgaard (1988), the amount of power possessed (discretion) coupled with the existence of a monopoly of power by the leadership and without accountability is a driving force for corruption. In this case, adequate quality of governance is needed to prevent corruption. Within the framework of the fraud triangle theory, an internal control system is needed as an effort to monitor and limit opportunities that cause fraud.
The results of previous studies show that the quality of good governance in public financial management can prevent and reduce the occurrence of corruption (Andersen, 2009; Al-Mahayreh & Abedelqader, 2015). Triwibowo (2019) provides empirical evidence showing that there is a long-term link between corruption and financial management quality. Thus, high quality of financial management can lead to a sound government. Improving the quality of good governance in government organizations should be carried out using various mechanisms, in particular building a strong internal control system to minimize the possibility of local officers behaving corruptly. The quality of governance needed to reduce corruption, among others, is in the form of an adequate internal control system. The role of the internal control system is very important to manage state finances in a way that is transparent, effective, and efficient. The greater the fiscal decentralization is (along with the implementation of a good internal control system), the less opportunity for corruption to occur in local governments. On the other hand, the worse the quality of governance, the greater the possibility of corruption.
Corruption can lead to low economic growth so it has a negative impact on GDP per capita, which in turn results in a low standard of living. In addition, corruption has an impact on less spending on health that leads to a lower level of life expectancy. Corruption also causes less spending on education so it has an impact on low human capital accumulation. Low level of standard of living, low level of life expectancy, and low level of human capital accumulation ultimately have an impact on the level of human development. The Human Development Index (HDI) is a comparative measure of life expectancy, literacy, education, and living standards. The HDI explains how the population can access development outcomes in terms of income, health, education, and so on. Akcay (2006) concludes that there is a significant negative effect between corruption and human development. Empirical evidence from the study shows that the higher the level of corruption in a country, the lower the level of human development. Bechererair and Tahtane (2017) examine the relationship between corruption and human development in Middle Eastern and North African countries with data from 1996 to 2012. It was found that there is a negative relationship between corruption and human development; the higher the level of corruption, the lower the quality of human development. Emara (2020) shows empirical evidence of corruption having a negative and significant impact on human development in both the short and long term. Urbina and Rodriguez (2021) also provide empirical evidence that corruption has a negative effect on human development.

AIMS AND HYPOTHESES DEVELOPMENT
This study aims to provide empirical evidence and analyze the effect of fiscal decentralization and the quality of local government governance on the level of corruption, as well as examine the impact of corruption on human development.
Based on the literature review, the following hypotheses are proposed: H1: Fiscal decentralization has a positive effect on the level of corruption.
H2: Poor quality of governance has a positive effect on the level of corruption.
H3: Corruption has a negative effect on the human development index.

METHODS
The population of this study is all local governments (districts/municipalities) on the island of Java, which number 113 organizations. This study uses the census method so that members of the population are used as samples. The local government on Java was chosen because it is the area with the most corruption cases compared to other islands in Indonesia. According to data from the Corruption Eradication Commission, there were 230 cases corruption cases in Java from 2014 to 2019, or 35% of corruption cases in Indonesia. The variable level of corruption in this study is measured by the number of rupiahs (IDR) in terms of state financial losses from corruption cases that already have legal force (inkrah) according to data on court decisions.
Fiscal decentralization is an unobserved variable measured by several proxies, namely from the revenue side seen from the amount of Regional Revenue (RR) and from the expenditure side it is seen from the General Allocation Fund (GAF), Special Allocation Fund (SAF), and Revenue Sharing Fund (RSF). The quality of governance is measured by the proxy of the internal control system. This proxy is measured by the number of findings of internal control weaknesses according to the results of the audit by the Supreme Audit Agency. Thus, the more weaknesses in the internal control, the worse the quality of governance.

RESULTS
Before testing the hypothesis, a descriptive analysis was carried out to describe the research data shown in Table 1 Note: RSF = Revenue Sharing Fund; GAF = General Allocation Fund; SAF = Special Allocation Fund; RI = Regional Income; ICS = Internal Control System; GDRP = Gross Domestic Regional Product; HDI = Human Development Index.
RSF, GAF, SAF, and RI. From the results of the output model (Table 2), it can be seen that the measurement model of the fiscal decentralization variable is acceptable, with a significant weight value for RSF, GAF, SAF, and RI having a p-value < 0.01 and a VIF value of each indicator < 3.3.
Evaluation of the structural model in this study was carried out by looking at the percentage of variance explained by looking at the Adjusted R-squared value, Goodness of Fit Model and q 2 predictive, effect size, and Full Collinearity VIF as well as the significance value of the path coefficient. From the results of the model fit output (Table 3), it can be seen that the model has a good fit, indicated by all indicators meeting the criteria in the rule of thumb (Kock, 2020).
After evaluating the measurement model and fit model, an analysis of the results of hypothesis testing was then carried out based on the Partial Least Squares-Structural Equation Modeling (PLS-SEM) estimation results. The results of the PLS-SEM test are presented in Figure 1.    Table 4.
The results of the hypothesis test in Table 4 show that fiscal decentralization has a positive effect on the level of corruption with a path coefficient of 0.100 and a significant 5% alpha. Thus, H1, which states that fiscal decentralization has a positive effect on the level of corruption, is supported. The results also show that the weakness of the internal control system has a positive effect on the level of corruption with a path coefficient of 0.062 and a significant level at 10% alpha. The results of this test show support for H2. The results in Table 4 also indicate that corruption in government organizations has a negative impact on the human development index. PLS-SEM test results show a negative path coefficient of -0.206 and significant at 0.01 alpha.
The results of this test show support for H3.

DISCUSSION
The results of PLS-SEM show that fiscal decentralization has a positive effect on the level of corruption. The higher the fiscal decentralization, the higher the level of corruption and vice versa; the lower the fiscal decentralization, the lower the level of corruption that occurs. The results of testing the first hypothesis in this study are in line with agency theory. The main characteristic of the relationship between the principal and agent is the existence of a delegation contract. Lupia and McCubbins (1994) state that such a contract emerges when one or several persons (or principal) choose another person/ persons to work following the needs and wishes of the principal. Fiscal decentralization has given great authority to local governments. Therefore, the powers, authorities, and responsibilities of district/ municipal governments become substantive.
The enormous authority in this era of fiscal decentralization has led to the emergence of high levels of discretionary power and monopolizing power. Therefore, the activities carried out by agents (local governments) cannot always be observed by the community (the principal) so there is information asymmetry. In addition, seen from the fraud triangle theory point of view, fiscal decentralization allows local governments to have more authority or autonomy in the decision-making process related to regional finance, thereby creating more opportunities for corruption. When the authority is becoming smaller, the opportunities for corruption are also becoming fewer.
The results of this study are consistent with several previous empirical studies, namely Rahayuningtyas  The results of PLS-SEM show that the weakness of the internal control system has a positive effect on the level of corruption. The worse the quality of governance -which is proxied by the more weaknesses in the internal control system -the greater the level of corruption in government organizations. This empirical evidence supports the fraud triangle theory that internal control weaknesses indicate weak monitoring and provide a greater opportunity for fraud to occur. The existence of authority and flexibility or freedom of local governments in managing, administering, or regulating regional finances -given by the central government -must be followed by strong supervision and control so that there are no deviations and irregularities. Strengthening the control function is carried out by implementing an internal control system. However, in this study, when viewed in terms of the descriptive statistics of the ICS variable, there are 238 districts and municipalities (52.65%) that have findings of ICS weaknesses above the average of 10.3. This indicates that the Note: * significant at alpha 10%, ** significant at alpha 5%, *** significant at alpha 1%.
district/municipality internal control system has not been effective. This empirical evidence supports by Triwibowo (2019) who found that misuse of governance causes the emergence of inconsistency cases at local authorities that is can be viewed as a corruption proxy.
The results of this study provide empirical evidence that corruption in government organizations has a negative effect on the human development index. The results of this study indicate that corruption is a serious problem in government organizations because it lowers the HDI. The empirical evidence provided by this study supports Akcay's (2006) argument that corruption can lead to low standards of living, low levels of life expectancy, and low levels of human capital accumulation, which in turn has an impact on low levels of human development. The results of this study also support the findings of Bechererair and Tahtane (2017), Emara (2020), and Urbina and Rodriguez (2021), who also provide empirical evidence that corruption has a negative effect on human development.

CONCLUSION
The purpose of this study was to examine the effect of fiscal decentralization and quality of governance on corruption in government organizations. In addition, this study also analyzes the effect of corruption on the level of human development. Based on the results of the study and the results of hypothesis testing, it can be concluded that fiscal decentralization has a positive effect on the level of corruption. This evidence confirms that the higher the regional authority to manage or regulate its finances, the higher the possibility of corruption. Poor governance, indicated by the many weaknesses in internal control as found by the Audit Agency, tends to lead to an increase in corruption. The results of this study also show empirical evidence that corruption is a serious problem because it can cause a decrease in the quality of human development. Since corruption has a negative impact on human development, fiscal decentralization must be managed properly with strong internal controls