The Effects of Local Government Characteristics and Audit Opinion on the Performance of District and City Governments in Indonesia

Article history: Received date: 10 December 2017 Received in revised form: 24 January 2018 Accepted: 18 April 2018 Available online: 30 September 2018 This study aims to examine the influence of district and city characteristics (fiscal decentralization ratio, dependency level on the central government, capital expenditure, age, regional status, population) and audit opinion on the performance of local and district government in Indonesia for the period of 20132015. Population of this study consists of all local government units of districts and cities in Indonesia for the years of 2013-2015. The purposive sampling method was used resulting in a total of 1,158 samples for three years period observations. The results of this study indicate that the variables of age area, population, and audit opinion have a positive effect on the performance of the studied Districts and Cities. Conversely, the dependency level on the central government has a negative effect on the performance of studied local governments. However, the ratio of fiscal decentralization, the capital expenditure, and the regional status have no effect on the performance of the local governments ©2018 FEB USK. All rights reserved.


Introduction
The Indonesian Law No. 32/2004 on district and city states that district and city manage their own governmental affairs according to the principle of autonomy and duty of assistance.The regional autonomy is implemented with the aim of accelerating the realization of community welfare and enhancing regional competitiveness.The existence of regional autonomy, people's aspirations will be more easily channeled and considered by policymakers to be a policy that stands for the interests of society (Artha, Basuki, & Alamsyah, 2015).The performance evaluation of district and city implementation is conducted in accordance to article 2 of Government Regulation Number 3 of 2007(Government Regulation No. 3/2007) on Reporting on district and city to a central government, and Government Regulation No. 6/2008 on evaluation guidelines for district and city.The results of the Performance Evaluation of District and City (PELG) of 2013-2015 issued by the ministry of home affairs of the district and city government in Indonesia shows that the performance of the district and city had improved very well.This is indicated by the increasing number of district and city governments that had achieved very high performance (Metrotvnews.com, 2017).
Among research related to the performance of district and city has been performed by Marfiana & Kurniasih (2013) and Mustikarini & Fitriasari (2012) who studied the characteristic of district and city and audit findings of Supreme Audit Agency.Meanwhile Ilmiyyah, Dewata, & Sarikadarwati (2017) tried to obtain empirical evidence that district and city finances correlated with audit opinions, legislative measures, intergovernmental revenue, size, leverage, and clusters of local financial capacity.This research is conducted to examine and obtain empirical evidence of factors that affect the performance of district and city administration.In contrast to previous research, this research uses the performance score of regional government performance in 2013-2015.This research aims to address the question whether ratio of fiscal decentralization, level of dependency on central government, capital expenditure, an age of district and city, regional status, population, and audit opinion influence both partially and simultaneously to the performance of district and city in Indonesia of 2013-2015.This research is expected to provide benefits as information in the framework of policymaking to improve the performance of district and city administration and as a reference for subsequent research on the performance of district and city administration.

Theoretical Framework and Hypothesis
Table 1 presents a summary of the differences in the results of the previous research related to the performance of district and city in Indonesia.LGR and CE had no effect on Y. (Juweny, 2016;Lesmana, 2010) 2011-2013 Provincial Government in Indonesia LGR had the positive effect on Y. GAF, RE, and OPINION had no effect on Y.
Note: ACTIVE = activity ratio; COMPLEX = regional complexity; CE = capital expenditure; EFFECTIVE = PAD effectiveness ratio; GAF = level of dependence on the Central Government; GROWTH = PAD growth ratio; LEVERAGE = loan; LGR = regional wealth and/or fiscal decentralization ratio; OPINION = audit opinion; POPULATION = number of locals; RE = regional expenditure; SIZE = area size; Y = performance of District and City administration.Artha et al. (2015) found that the size of the region had a significant positive effect on the performance of the district and city in the province of West Nusa Tenggara in Indonesia.Sedyaningsih & Zaky (2015) similarly found that the level of dependence on the central government and BPK findings had a significant negative effect on the performance of district and city in South Sulawesi.Meanwhile Juweny (2016) and Sudarsana & Rahardjo (2013) identified that the fiscal decentralization ratio positively affects the performance of district and city in Indonesia.Rustiyaningsih, Sri, & Immanuela (2014) proved that the number of the population affects the performance score of district and city.

The Effect of Ratio of Fiscal Decentralization on the Performance of District and City
According to Mahmudi (2016), the degree of decentralization was calculated based on the ratio between the amount of Locally-Generated Revenue to total local revenue.This ratio indicated the contribution of LGR (Locally-Generated Revenue) to total local revenue.The higher the contribution of LGR, the higher the ability of district and city decentralization.Positive growth encouraged investment hence it simultaneously encourage the improvement of local infrastructure.The quality of public service would reflect the performance of a district and city.This was supported by the research results of Handayani & Dwi (2016); Juweny (2016); Mustikarini & Fitriasari (2012);and Qowi & Prabowo (2017) who found that fiscal decentralization ratio positively affected the performance of district and city in Indonesia.Therefore, it is hypothesized that: H1: Fiscal decentralization ratio positively affects the performance of district and city administrations.

The Effect of Dependency on Central Government on Performance of District and City
According to Mahmudi (2016), the ratio of local dependence is calculated by comparing the amount of transfer revenue received by the district and city with total local revenue.Based on Law No.33/2004, Balancing funds are funds sourced from state budget revenues allocated to the region to fund the needs of the region in the context of implementing decentralization; and the General Allocation Fund (GAF) are funds sourced from state budget revenues allocated for the purpose of equitable financial capacity among the regions to fund the needs of the region in the context of the implementation of decentralization.The higher this ratio, the greater the level of dependence of the district and city on the central government and/or provincial government.The greater the GAF revenue for one region, the government would better monitor the implementation of the GAF allocation compared to the areas with fewer receipts.This could motivate the district and city to perform better because the supervision of the central government is greater.Thus, the higher the GAF from the central government is expected to improve the service of district and city to the community so that the performance of district and city is also improving (Sudarsana & Rahardjo, 2013).This was supported by the results of research Mustikarini & Fitriasari (2012) which revealed that the level of dependence on the central government had a positive effect on the performance of district/city government.Thus the hypothesis in this study is: H2: Dependency on central government has a positive effect on the performance of district and city administration.

The Effect of Capital Expenditure on Performance of District and City
Large capital expenditure is a reflection of the number of infrastructure and facilities built.So that more development would improve service to the community so that the performance of the region will be better (Sudarsana & Rahardjo, 2013).According to Law No.32/2004 Article 167 paragraph 1, regional expenditure is used to protect and improve the quality of community life in an effort to fulfill local obligations.This is manifested in the form of improved mandatory and optional services, such as basic services in education, provision of health care facilities, social facilities, decent public facilities, and developing social security systems.So it can be concluded that the higher spending of district and city should reflect the higher level of service provided to the community.Furthermore, the higher the level of service provided, the higher the performance score of district and city (Mustikarini & Fitriasari, 2012).According to Sumarjo (2010) that development expenditure positively affected the financial performance of district and city in Indonesia.However, Sedyaningsih & Zaky ( 2015) stated that the level of dependence on the central government has negatively affected the performance of the regency government in South Sulawesi.From the above description, the hypothesis of this study is: H3: Capital expenditure has a positive effect on the performance of district and city administration.

The Effect of District and City's Age on Performance of District and City
The age of district and city can be interpreted by how long the district and city exist (Mandasari, 2009).Meanwhile, according to Setyaningrum & Syafitri (2012), the administrative age of the district and city is the year of the establishment of a district and city that is officially established in the law on the establishment of the area.The longer the existence of a district and city, the experience is better understood than the new district and city.Likewise, with the administrative system, district and city with older ages with various experiences will have better administrative and record keeping processes (Lesmana, 2010).District and city with a longer administrative age would be more experienced and had a better ability to manage their area and organize their government.This was because the district and city that have a longer life had often conducted an evaluation of the administrative and financial reports.So that evaluation results would be followed up to improve the performance of the district and city.
Based on the description, it could be said that older district and city or those with longer lives would have better performance than younger district and city.This was supported by the results of research by Lesmana (2010) and Setyaningrum & Syafitri (2012) which proved that the age of government had a significant effect and had a significant positive effect on the level of financial statement disclosure.Thus, it is hypothesized that: H4: The age of district and city has a positive effect on the performance of the district and city administration.

The Effect of Local Status on Performance of District and City
According to article 1 of Law No.18/1945 (I), the state recognizes and respects district and city units that are of a special or special nature regulated by law.This type of area affects the completeness of disclosure due to differences in community characteristics and income structures have implications for different social controls.Therefore, this research hypothesis is: H5: Local status has a positive effect on the performance of the district and city administration.

The Effect of Population on Performance of District and City
According to the Indonesia dictionary online (http://kbbi.web.id),residents are people who inhabit a place (village, city, island, etc.).The number of people in a region will affect the services provided by the district and city.The greater the population of a region, the higher the public demand for the district and city to improve the better public services.With that demands the government would be encouraged to improve performance in providing services to the community.This was supported by the results of research Rustiyaningsih et al. ( 2014) which proved that the number of population affected the performance of district and city of district/city.Therefore, this research hypothesis is: H6: The number of the population positively affects the performance of the district and city administrations.

The Effect of Audit Opinion on Performance Score of District and City
According to article 1 of Law No.15/2004, opinion is a professional statement as the auditor's conclusion about the fairness of the information presented in the financial statements.There are four opinions given by the examiner, namely Unqualified Opinion (UO), Qualified Opinion (QO), Adverse Opinion (AO) and Disclaimer Opinion (DO).Thus, the better the opinion was given by the Supreme Audit Agency, the better the performance of the district and city.This was supported by the results of research Handayani & Dwi (2016) which proved that the audit opinion had an effect on the performance of district and city of district/city in Indonesia.From the above description, then the hypothesis in this study are: H7: Audit opinion has a positive effect on the performance of the district and city administration.Here is a framework that describes the research model and the relationship between variables used in research:

Research Method
According to the type of data and analysis, this research is a quantitative research.This study uses data in the form of numbers or qualitative data that was incurred (quantitative data).Meanwhile, based on the level of explanation (level of clarity), this research was classified as associative research because it showed the suspicion about the relationship between two or more variables with other words aimed to determine the influence of one variable on other variables (Sugiyono, 2016).

Population and Sample
Population in this research is all government of district and city in Indonesia of 2013-2015.While the sample in this research were 1158 district and city government in Indonesia.Sample selection technique in this research was purposively sampling with the criterion of district and city government having scores and status of performance evaluation of district and city of 2013-2015, and had complete data in the form of budget realization report, audit opinion, population, and age of area.In this study, related to district and city, researchers issued the province as a sample of research because districts and cities have the same authority.The district and the city are autonomous regions that are authorized to regulate and manage their own government affairs.The data used in this study was the audited district and city financial Statements obtained from the Supreme Audit Agency, and the audit opinion variable data obtained from the website of the Supreme Audit Agency of the Republic of Indonesia (www.bpk.go.id).The data of district and city's ages and ratings of district and city government performance scores are obtained through the website of the Directorate General of Regional Autonomy (www.otda.kemendagri.go.id) as well as population data obtained through the website of the Directorate General of Population and Civil Registration (www.dukcapil.kemendagri.go.id).

Results And Discussion
The descriptive statistics indicated that 697 districts and city (60.18%) had above average performance scores and remaining 461 districts and city (39.82%) had below average performance scores.This means that during the year 2013-2015 there were many district and city who had high-performance scores.Figure 3    Before the hypothesis testing, the research model was tested to find any violations of classical assumptions ranging from data normality, multicollinearity, and autocorrelation.there were no violations of classical assumptions in the research model.Furthermore, table 4 contained a summary of the test results of the research model, ie.coefficient, t-statistic, probability for each independent variable, R square value and F value and significance.To examine the effect the ratio of fiscal decentralization, level of dependency on central government, capital expenditure, an age of district and city, regional status, population, and audit opinion on the performance of district and city in Indonesia, it is used multiple linear regression statistical analysis.The analysis is processed with the application program eViews 9.In table 4 shows the result that the value of R square was 0.525.This meant that the variation in district and city performance could be explained by the variables of the decentralization ratio, the dependency ratio on the central government, the ratio of capital expenditure, the age of the district and city, the local status, the total population, and the audit opinion by 52.5%.While the rest of 47.5% influenced by other variables that are not included in this research model.
F value counted 17.182, while F table value of 2.01 with sample amounting to 1158 (df = ∞), independent variable of 7 (k = 7).This means independent variables consisting of decentralization ratios, dependency ratio on the central government, capital expenditure ratios, the age of district and city, local status, population, and audit opinion had the significant simultaneous effect on the performance of district and city government in Indonesia.

The Effect of Fiscal Decentralization Ratio to Performance of District and City
Based on table 4, there is the t value 1.498 <t table 1.962 and probability values 0.401 > 0.05.This means that hypothesis 1 was rejected, the fiscal decentralization ratio had no significant effect on the performance of the district and city government.The coefficient value for this variable showed a value of 0.671.The positive sign indicated a unidirectional relationship between the fiscal decentralization ratio (independent variable) and the performance score of the district and city government (dependent variable).The results of this study was in line with research conducted Artha et al. (2015) and Sedyaningsih & Zaky (2015) However, it was not in line with the results of research conducted by Harumiati (2015); Juweny (2016); Mustikarini & Fitriasari (2012);Rustiyaningsih et al. (2014);and Sudarsana & Rahardjo (2013) where the ratio of fiscal decentralization or local wealth had a positive effect on the performance score of district and city.
The results of this study suggested that district and city with large LGR would have high-performance scores as well.If the district and city had a large LGR that had a lowperformance score then the district and city should evaluate its performance, which meant that its performance was not better than the district and city which had a smaller LGR.district and city with large LGR were required to better manage and utilize the wealth it had for the sake of service to the community (Sudarsana & Rahardjo, 2013).
Judging from the average value of fiscal decentralization ratio of 0.0977 or 9.77% indicated that the share of LGR to total income was still very low.This meant that the district and city governments in Indonesia were not good enough in optimizing or exploring the potential wealth of their respective regions to meet the needs of their own regions, as well as the district and city governments were still heavily dependent on the central government in the administration of government affairs.

The Effect of Dependency on Central Government on Performance of District and City
Based on table 4, there is the t value 3.024 > t table 1.962 and the probability value of 0.005 < 0.05.This showed hypothesis 2 was accepted, this meant that the dependency level on the central government had a significant effect on the performance score of the district and city and the city administration.The coefficient value for this variable showed a value of -0.462.In contrast to the fiscal decentralization ratio, the coefficient value for this variable indicated a negative sign which implied an opposite relationship between the level of dependence on central government (independent variables) and the performance of district and city administrations (dependent variables).
Means the greater the level of dependence on the central government then the performance of district and city and district governments will be lower and vice versa.The results of this study was in line with the research conducted by Sedyaningsih & Zaky (2015), but not in line with the results of research conducted Marfiana & Kurniasih (2013) and Mustikarini & Fitriasari (2012) that the level of dependence on the central government had a positive effect on the performance score of district and city.On the contrary, the results of research by Artha et al. (2015); Harumiati (2015); Juweny (2016);Rustiyaningsih et al. (2014);and Sudarsana & Rahardjo, (2013) stated that the level of dependence on the central government did not affect the performance score of district and city.
Judging from the average value of dependence on the central government (GAF) of 55.89%, it showed that the share of GAF to total income was quite large because more than half of the total regional income came from the GAF.This meant that the district and city governments in Indonesia were still very dependent on the central government to meet the needs of the regions in the implementation of their respective governmental affairs.From the value of the level of dependence on the central government (GAF), the GAF also has a significant role in the implementation of district and city in the district and city governments in Indonesia as evidenced by the coefficient value of -0.462, which was the greatest value compared to other variables.But with the amount of revenue obtained, not necessarily the funds were also used to improve the performance of district and city that could be seen with the development in the area, in other words, General Allocation Fund is used for other needs such as operational expenditure.This was supported by data showing that the average value of the capital expenditure to total expenditure was still low at 24.70%.

The Effect of Capital Expenditure on Performance Score of District and City
Based on table 4, there is the t value of 1.672 <t table 1.962 and the probability value of 0.093> 0.05.This meant that hypothesis 3 was rejected.The capital expenditure ratio had no significant effect on the performance of district and city government.The coefficient value for this variable showed a value of -0.378.The negative sign indicated the opposite relationship between the ratio of capital expenditure (independent variable) and the performance of the district and city government (dependent variable).The results of this study was in line with research conducted by (Artha et al., 2015;Sedyaningsih & Zaky, 2015;Sudarsana & Rahardjo, 2013), but not in line with the results of research conducted by (Marfiana & Kurniasih, 2013) that capital expenditure had a positive effect on the performance of district and city.
The results of this study supported the assertion that there was information asymmetry between the agent (government) that had direct access to capital expenditure with the principal (community) (Sudarsana & Rahardjo, 2013).The results of this study had implications for district and city district and city that large spending should be able to provide good performance.However, these findings proved that capital expenditures such as infrastructure costs had not been able to improve the performance of district and city to the community.District and city should issue policies in managing capital expenditures that could improve regional performance to the public.

The Effect of District and City's Age on Performance of District and City
Based on Table 4, there are the t value of 2.276> t able 1,962 and the probability value of 0.041 <0,05.This means hypothesis 4 was accepted.The age of district and city had a significant positive effect on the performance of the district and city government.
The coefficient value for this variable showed a value of 0.001.In contrast to the level of dependence on central government and capital expenditure ratios, the coefficient value for the regional age variable showed a positive sign which indicated a unidirectional relationship between the age of the region (independent variable) with the performance of the district and city government (dependent).It means the longer a district and city was formed then the performance of the district and city would be higher.The results of this study were in line with the research of Setyaningrum & Syafitri (2012) and Lesmana (2010) where the older district and city formed had an influence on the wider level of mandatory disclosure of financial statements.

The Effect of Local Status on Performance of District and City
Based on table 4, there are the t value is 1.829 < t table 1.962 and the probability value of 0.634 > 0.05.This meant hypothesis 5 was rejected.Local Status had no significant effect on the performance of the district and city governments.The results of this study were in line with research conducted by Harumiati (2015).This indicated that whatever the status of the region, whether district or city would not affect the performance of district and city, and contrary to the results of research Pranataningrum (2012) which states that the complexity of the region positively affects the performance of district and city.

The Effect of Population on Performance of District and City
Based on table 4, there are the t value of 3.987 > t table 1,962 and probability value of 0,050 > 0,05.This meant hypothesis 6 was accepted.This meant that the number of residents had a significant effect on the performance scores of the district and city.The coefficient value for the population variable was 0.079 and showed a positive sign which indicated a unidirectional relationship between the population (independent variable) and the performance of the district and city government (dependent variable).It meant the more residents number in a district and city then the performance of the district and city of the district and city would be higher.The results of this study were in line with Rustiyaningsih et al. (2014) research.This showed that the greater the population of a region, the higher the public demand for the district and city to improve the better public services.

The Effect of Audit Opinion on Performance of District and City
Based on table 4, there is the t value of 2.458 > t table 1.962 and the probability value of 0.023 < 0.05.This meant hypothesis 7 was accepted.Audit opinion had a significant positive effect on the performance score of the district and city government.The coefficient value for the audit opinion variable was 0.275 and showed a positive sign which indicated a unidirectional relationship between the audit opinion (independent variable) and the performance score of the district and city administration (dependent variable).It meant that the better the audit opinion, the performance score of the regional government and the city would be higher.The results of this study had implications for district and city policy to be more careful related to the management of local finances.Audit opinion had relevance to district and city financial performance meant that the audit opinion of a good district and city was perceived as an indicating of a good financial performance too.The results of this study were in line with Sutopo, Wulandari, Adiati, & Saputra (2017) which shows the audit opinion has a positive effect on the performance of local government implementation and means audit opinion can be used as an indicator of the performance of local government, but the results of this study was not in line with the results of research conducted by Ilmiyyah et al. (2017) and Marfiana & Kurniasih (2013) that audit opinion did not affect the financial performance of district and city.Table 4 shows the F value of 17.182 > F table value of 2.01.Thus hypothesis 8 was accepted.This meant that independent variables consisting of decentralization ratios, dependency ratio on the central government, capital expenditure ratio, district and city's age, local status, total population, and audit opinion.If implemented together will give a positive significant effect on the performance of district and city in Indonesia year 2013-2015 with F-Sig 0,000 and the value of R square was 52.5%.

Conclusions, Limitations, and Suggestions
The findings in this study indicate that fiscal decentralization ratio, the level of dependence on the central government, capital expenditure, the age of local government, local status, population, and the audit opinion had a significant effect on the performance of district and city in Indonesia of 2013 -2015 Partially, the age of district and city, population, and audit opinion had a positive significant effect on the performance of district and city in Indonesia of 2013-2015.Conversely, the dependency on the central government had a significant negative effect on the performance score of district and city in Indonesia of 2013-2015.However, the ratio of fiscal decentralization, capital expenditure, and local status had no significant effect on the performance of district and city in Indonesia of 2013-2015.This main limitation of this research is because the period used is only 3 years so it cannot be used to see the consistency of results in the long run.Based on this limitation, future research is suggested to extend observation period to see consistency of research results.In addition, researchers may conduct also observe what factors that cause the non-significant effect of the ratio of fiscal decentralization, capital expenditure, and local status on the performance of district and city in Indonesia.
Figure 1 Research Framework Figure 3The percentage of District and City Performance Score Based on figure 4 below, the average score of performance scores of district and city governments from 2013-2015 was still on the high achievement criteria although there had been an increase in the average performance score of the district and city from 2013-2015.In 2013 the average performance score was at 2.360003, while in 2014 there was an increase of 9.42% to 2.605700 and by 2015 the average performance score also increased by 8.56% to 2,849522.The following figure 4.2 is presented the average score of Performance Evaluation of District and City (PELG) of 2013-2015: Source: Data researcher, 2017.

Figure 4
Figure 4 Average Performance Evaluation of District and City (PELG) of 2013-2015

Table 1
Summary of Previous Research Findings

Table 4
Multiple Linear Regression Test Results