Local Government Characteristics, Adoption of Accrual Accounting, and Government Performance

Accrual accounting is the accounting basis that began to be applied to Indonesian Government Accounting Standards (SAP) in 2015. This study aimed to examine the effect of local government characteristics on the financial performance of local government using the adoption of accrual-based Government Accounting Standard as a mediation variable. The sample is the local government in the Provinces of DIY and Central Java. This research was analyzed by Partial Least Square (PLS). The results showed that leadership education background, experience using the accrual basis, complexity, and local government’s wealth have a positive effect on the adoption of accrual-based Government Accounting Standard. Furthermore, leadership education background, and experience using accrual basis have a positive effect on the financial performance of local government. In contrary, the adoption of accrualbased Government Accounting Standard and local government’s expense have a negative effect on the financial performance of local government. The adoption of accrual-based Government Accounting Standard mediates the effect of leadership education background, experience using accrual basis, and local government’s wealth on the financial performance of local government.


Introduction
Local government is a public sector organization that uses public sector accounting information for making decisions.Accrual accounting is the basic accounting that is currently being applied to Government Accounting Standards in Indonesia. Previously, Indonesia has implemented cash-based accounting and cash-based accounting toward accruals. Some previous research states that there is a need to use accrual accounting.
Government Regulation No. 71 of 2010 is a government regulation that explains about Government Accounting Standards. This government regulation was issued on October 22, 2010. The application of accrual-based accounting will be carried out in stages. Local government implemented cash-based towards accruals until 2014. It is expected that all local government in Indonesia have used the preparation of financial statements on an accrual basis in 2015.
Local government performance can be seen from the economy, efficiency and effectiveness (Sumenge, 2013). According to the 2009 Semester Examination Results by the Audit Board of the Republic of Indonesia, 2011Indonesia, , 2014 and 2016, the Audit Board of the Republic of Indonesia found non-compliance with statutory provisions. This resulted in waste in Indonesia decreasing from the year before the issuance of Government Regulation No. 71 of 2010 when compared to the year after accrual-based accounting standards were issued. In addition, the Highlights of Semester Examination Results by the Audit Board of the Republic of Indonesia show that the development of opinion is getting better in 2014 and 2016. The Unqualified Opinions obtained by Local Government Financial Statements in Indonesia have increased since the issuance of Government Regulation No. 71 of 2010. Patrick (2007) found that the characteristics of local government have a positive association with the application of innovation.
The contribution is this study using mediation variables on the accrual-based Government Accounting Standard adoption in the local government financial statements. The researcher used this mediation variable because the adoption of accrual-based Government Accounting Standard in Indonesia was only implemented about 2 years ago or about 5 years after the issuance of Government Regulation No. 71 of 2010. Accrual-based Government Accounting Standard is a variable that is influenced and influences other variables, such as in the research of Beechy (2007) and Patrick (2007). In addition, this study uses the private sector financial ratios that have been conducted by Cohen S. (2008a).

Literature Review
The introduction of accrual accounting is based on institutional theory and new public management (NPM). Organizations must adapt to social norms of behavior that can be accepted by the community so that the organization can maintain its survival. In addition, one formation of institutional theory is coersifisomorphis. It suggests that organizations adopt other organizations because of the pressures of the state and other organizations or the wider society (DiMaggio and Powell, 2000). The accrual-based Government Accounting Standard can record more flexible financial transactions. Accrual-based Government Accounting Standard is issued by the government to adjust government accounting transactions that are increasingly complex. Meanwhile, the government as an agency applies the functions and authorities of the government that exist in institutional theory to carry out the rules set by the accrual-based Government Accounting Standard.
NPM assumes that private sector management practices are better than public sector management practices. NPM gives greater attention to the achievement of the performance and accountability of public managers. In addition, NPM wants a more flexible modern organization model (Hood, 1995). The accrual-based Government Accounting Standard is adopted from accounting standards applied by the private sector. The adoption of accrualbased Government Accounting Standard is based on the principles of NPM concept. It is expected to improve local government performance through accountability and transparency.
The local government must implement accrual-based Government Accounting Standard as a rule that must be obeyed. It is because of the principle of institutional theory which states that an organization should be able to adapt based on acceptable norms. Thus, the NPM concept emphasizes the principle applied in accrual-based Government Accounting Standard. The function and authority of the government as an institution that runs accrual-based Government Accounting Standard has been described in institutional theory.
Accrual-based Government Accounting Standard has PSAP 12 regarding Operational Reports. Operational reports provide information on government operational activities. Entity reports transparently the amount of economic resources obtained and the amount of expense to carry out government activities. Components of government financial statements, consist of Budget Realization Report, Report on Changes in Budget Balance More, Balance Sheet, Operational Report, Cash Flow Statement, Statement of Changes in Equity, and Notes of Financial Report. The accounting basis used in government financial statements is the accrual-based for recognition of income on Operational Report, expenses, assets, liabilities, and equity. The budget is prepared and implemented on a cash basis. Therefore, the Budget Realization Report is prepared on a cash basis. Paulsson (2006) conducted a study entitled Accrual Accounting in Public Sector: Experiences from the Central Government. Paulsson (2006) examined the presentation of experiences from the use of accrual accounting information in the central government in Sweden. Accrual accounting information has little benefit to public organizations. Data analysis showed that do not provide a full view of the usefulness of accrual accounting information in the Swedish central government. Accrual accounting information is used for a relatively broad range, but it is difficult to link these uses to the budget process or to specific decisions. Information is used as the main source of information in different situations, whether in agent offices or government offices. (Becker et al., 2014;Falkman and Tagesson, 2008;Gårseth-Nesbakk, 2011;Harun et al., 2012;Mu'am, 2015;Widyanti, 2015;Zawitri et al., 2015;Zelmiyanti, 2015) examined the application of accrual-based accounting that influences better practices, such as compliance with more standards well because the information generated by accrual-based accounting was more compliant with GAAP. In addition, accrual accounting in the public sector economy provided ideas for politics and power to provide information about the complexity of institutionalization in a political-economic environment. Based on the results of previous studies, the proposed hypothesis is: H1: The adoption of accrual-based Government Accounting Standard has a positive effect on the local government financial performance. Windels and Christiaens (2008) conducted research on active involvement, experience using accounting, public interest, support from accounting consultants, support from the board, support from IT consultants, special training, support from internal audits, local city support. The study, entitled The Adoption of Accrual Accounting in Flemish Public Centers For Social Walfare: Examining The Importance Agents of Change states that de facto implementation requirements are limited and strongly influenced by the interests of local management and the involvement of consultants. The results of the study stated that professional support and consultation with consultants affected the application of accrual accounting. Beechy (2007) conducted research on Non-Profit Organizations, private goods, government goods, and full accrual accounting as research variables of Non-Profit Organization (NPO), Private Goods, Public Goods, Full Accrual Accounting under the title Does Full Accrual Accounting Enhance Accountability? The results of the study stated that full accrual accounting was not a single concept. Full accrual accounting consisted of accrual accounting, expense recognition, interpriode allocation. Every organization should use accrual accounting, but the ability to apply expense recognition and inter-period allocation depends on the characteristics or nature of the organization. (Azmi and Mohamed, 2014;Badriyah, 2015;Becker et al., 2014;Eriotis et al., 2011;Langelo et al., 2015;Maimunah, 2016;Tarigan and Nurtanzila, 2013;Windels and Christiaens, 2008) examined the readiness of organizations in the implementation of accrual-based accounting. The results of the study stated that technology preparedness, quality of human resources, time needed in implementation, procurement of socialization, staff training, and technical guidance in the implementation of accrual-based Government Accounting Standard. Patrick (2007) explained that there are several factors that influence organizational innovation in implementing GASB 34 in the Pennsylvania local government. GASB 34 calls on many local governments to apply modifications to accrual basis accounting and record the assets of retroactive assets. GASB 34 aimed to improve transparency and accountability of the government.
H2: The leadership education background has a positive effect on the adoption of accrual-based Government Accounting Standard.
H3: The experience using an accrual basis has a positive effect on the adoption of accrual-based Government Accounting Standard.
H4: The complexity has a positive effect on the adoption of accrual-based Government Accounting Standard.

H5:
The leadership education background has a positive effect on the financial performance of the local government.
H6: The Experience using an accrual basis has a positive effect on the financial performance of the local government.
H7: The complexity of local government has a positive effect on local government financial performance. H8: The adoption of accrual-based Government Accounting Standard mediates the influence of the leadership educational background on the local government financial performance.
H9: The adoption of accrual-based Government Accounting Standard mediates the effect of experience using the accrual basis and local government financial performance.
H10: The adoption of accrual-based Government Accounting Standard mediates the effect of local government complexity on local government financial performance.
Cohen Sandra (2008b) conducted a study of profitability ratios, liquidity ratios, capital structure ratios, performance ratios, macroeconomic factors such as wealth and regional size as research variables. The title of the research carried out was the results of the study which stated that the Profitability Ratio, Liquidity Ratio, Capital Structure Ratio, Performance Ratio, Macro-Economic Factors Selection, such as wealth and size. The results of the study stated that the assessment of financial performance using traditional financial ratios was influenced by economic factors, namely GDP, population, real estate value, tourist development, and size. Real estate prices have a significant positive effect on profitability ratios. H11: The local government's wealth has a positive effect on the adoption of accrual-based Government Accounting Standard.
H12:Thelocal Government's wealth has a positive effect on the financial performance of the local government. H13: The local government's expense has a positive effect on the adoption of accrual-based Government Accounting Standard.
H14: The local government's expense has a negative effect on the financial performance of local government. H15:The adoption of accrual-based Government Accounting Standard mediates the influence of local government's wealth on the financial performance of local government.
H16: The adoption of accrual-based Government Accounting Standard mediates the impact of local government's expense on local government financial performance.

Figure-2. Research Model
The research model explains that local government characteristics is related to the adoption of government accounting standard. In other that, the adoption of government accounting standard is related to the financial performance of local government. In addition, the better local government characteristics make it more possible to adoption of government accounting standard and the better the adoption of local government accounting standard makes it more possible to financial performance of local government.

Methodology/Materials
Data was collected from the Audit Report and Local Government Financial Statements for 2015 to 2016 which was prepared based on the Statement of Government Accounting Standards. Data was from complete Audit Report and Local Government Financial Statements consisting of Audit Report and Local Government Financial Reports, Internal Control Systems and Compliance Audit Report to Laws 2015-2016 obtained directly from Audit Board of Daerah Istimewa Yogyakarta and Audit Board of Central Java. Daerah Istimewa Yogyakarta consists of 4 districts and 1 city. Central Java consisted of 29 regencies and 6 cities.

Dependent Variables
Local government financial performance used in this study is the Surplus on Assets (SOA) ratio which is a comparison between surplus and total assets in a certain period. SOA = (Net Surplus or Deficit) / (Total Assets)

Mediation Variables
The mediation variable in this study is the adoption of accrual-based Government Accounting Standards. This variable measurement uses elements of accounting index (Windels & Christiaens, 2007) that are adapted to the conditions in Indonesia based on Government Accounting Standard and the data can be accessed and available in the Local Government Financial Report (LKPD). Each element is measured in a dichotomy. Each item disclosed based on the table accounting index is given a value of 1 and undisclosed items are given a value of 0. The table accounting index that has been adjusted to Indonesia's conditions is listed in Table 2.2 The adoption accrual-based Government Accounting Standard = (Number of items disclosed) / (Total number of items) x 100%

Independent Variables
The independent variable in this study is the characteristics of the local. Non-financial characteristics of local government in this study used three independent variables, namely leadership education background, experience using accrual basis, and complexity of local government. Leadership education background (LBP) is measured using dichotomous variables. If the head of the work unit comes from an economics, accounting or management graduate, then it is given a value of 1 and if it is given a value of 0. The experience of using accrual-based Government Accounting Standard is measured by using the number of years since the first time the work unit uses the accrual basis in its financial statements. The complexity is measured by the amount of the local work units in a local government. Financial characteristic of Local government has two variables, namely the level of wealth and expense of local government. Local Government Wealth measured using total assets. Local government expenses are measured using the amount of expense on the operational report.

Results and Findings
The method of analysis consisted of descriptive statistics and inferential analysis using Partial Least Square (PLS). All indicators in each variable have the largest cross loading value on the variable that is formed, which is worth 1.  BLO is an operating expense measured using the amount of expense on the operational report. LBP is a leadership educational background that is measured using a dichotomy scale, 1 for economic education background, 0 for noneconomics. PBA is an experience using an accrual basis measured using the number of years of LG experience using the accrual basis. S OA is the financial performance of local government, iesurplus on assets as measured by comparing the amount of surplus or deficit to total assets. SAP is the adoption of accrual-based SAP as measured by comparing the number of items disclosed against the total number of items used. SKP is the complexity measured by the amount of SKPD. TAN is a wealth that is measured using total assets.
The value of composite reliability, cronbach's alpha, and average variance extracted each indicator on leadership background variables, experience using accrual basis, and local government wealth, and financial local government is > 0.7. DFS is a leadership educational background that is measured using a dichotomy scale, 1 for the r rata educational economy, 0 for non economics. PBA is an experience using an accrual basis measured using the number of years of local government experience using the accrual basis. SKP is the complexity measured by the amount of local work units. SAP is the adoption of accrual-based Government Accounting Standard as measured by comparing the number of items disclosed against the total number of items used. TAN is a wealth that is measured using total assets. BLO is an operating expense measured using the amount of expense on the operational report. SOA is the financial performance of local government, iesurplus on assets as measured by comparing the amount of surplus or deficit to total assets. LBP is a leadership educational background that is measured using a dichotomy scale, 1 for economic education background, 0 for non economics. PBA is an experience using an accrual basis measured using the number of years of local government experience using the accrual basis. SKP is the complexity measured by the amount of local work units. SAP is the adoption of the adoption of accrual-based Government Accounting Standard as measured by comparing the number of items disclosed against the total number of items used. TAN is a wealth that is measured using total assets. BLO is an operating expense measured using the amount of expense on the operational report. SOA is the financial performance of local government, iesurplus on assets as measured by comparing the amount of surplus or deficit to total assets. LBP is a leadership educational background that is measured using dichotomy scale, 1 for economic education background, 0 for non economy. PBA is an experience using an accrual basis measured using the number of years of local government experience using the accrual basis. SKP is the complexity measured by the amount of SKPD. SAP is the adoption of accrual-based SAP as measured by comparing the number of items disclosed against the total number of items used. TAN is a wealth that is measured using total assets. BLO is an operating expense measured using the amount of expense on the operational report. SOA is the financial performance of local government, iesurplus on assets as measured by comparing the amount of surplus or deficit to total assets. The first hypothesis test results showed that the adoption of accrual-based Government Accounting Standard had a negative and significant effect on the adoption of accrual-based Government Accounting Standard with p-value value of 0.000<0.05. Hypothesis test results did not support the first hypothesis. The results of this study were in line with the results of Cohen S. (2008a) which stated that the adoption of accrual-based accounting had a negative effect on surplus on assets (SOA).The accrual-based Government Accounting Standard did not record the payment of tax receivables last year as income this year, but records are made when the tax receivables occur. The recording method also applied to recording costs. This methode results in recording smaller operating income and recording expenses tends to be greater in operational reports. Therefore, the surplus generated will tend to be smaller so that the SOA produced also tends to be smaller. So, the application of accrual-based Government Accounting Standard caused SOA to be smaller.
The second hypothesis test showed that the educational background of the leadership had a positive and significant effect on the adoption accrual-based Government Accounting Standard with p-value value of 0.043 <0.050. Hypothesis test results supported the second hypothesis. The results of this study were supported by the results of research from (Tarigan and Nurtanzila, 2013) which stated that the use of accrual basis should consider the government's readiness as executive and parliament as the people's representatives in overseeing the current government program. However, this study was not in line with Usman et al. (2016) which stated that there was no significant effect from the educational background of the leadership. Research from Eriotis et al. (2011) also stated that the educational background of the CEO had no significant effect on the adoption of accrual accounting in Greek government hospitals.
The third hypothesis test showed that experience on an accrual basis and significant positive effect on the adoption of accrual-based Government Accounting Standard with value p-value 0.000 <0.050. Hypothesis test results supported the third hypothesis. The results of this study were not in line with Usman et al. (2016) research and the study of Eriotis et al. (2011) stating that there was no significant effect of the experience using the accrual basis for applying the accrual accounting application to the public sector. In addition, the results of this study were not in line with research from (Windels and Christiaens, 2008) which stated that prior experience using accounting, active engagement, and support from consultants had no significant effect on the application of accrual accounting.
The 4 th hypothesis test results showed that the complexity of local government did not significantly influence the adoption of accrual-based Government Accounting Standard with p-value value of 0138 <0.050. Hypothesis test results did not support the fourth hypothesis of this study. The complexity of local government in this study used the number of local work units. Indicator of local work units amount did not show Human Resource capability in the adoption of accrual-based Government Accounting Standard. Langelo et al. (2015) stated that the number of implementers was still not enough in each work unit and the readiness of supporting devices was not tested. Therefore, improving the quality and amount of human resources with an accounting education background and technical guidance are very necessary. In addition, Widyanti (2015) stated that the implementation of accrual-based accounting has faced many obstacles, such as the lack of human resource capacity of the government apparatus, incorrect placement of human resources, and a lack of understanding of the state financial administration staff.
The 5 th hypothesis test results showed that the leadership educational background had positive and significant effect to the level of financial performance of local government with p-value of 0.000 <0.050. Hypothesis test results supported the hypothesis. Patrick (2007) stated that individuals and teams made decisions within the organization, but those decisions were influenced by the behavior and beliefs of top management that will then influence the members of the organization. The result of 6 th hypothesis test showed that experience using accrual basis had no significant effect to local government performance with p-value value is 0.636> 0.050. Hypothesis test results did not support the hypothesis. Local government performance measurement would be more appropriate if using accounts in the financial statements and using the ability of human resources in managing local finances such as Cohen S. (2008a).
The 7 th hypothesis test results showed that the complexity of local government did not have a significant effect on the financial performance of local government in this study with p-value of 0.187> 0.050. Hypothesis test results did not support the hypothesis. The complexity of local government used the number of local work units in this study. According to Patrick (2007) complexity was the level of specialization of a job or a person's expertise in a job. Job specialties were usually measured using the formal knowledge of organizational members.
The 11 th hypothesis test showed that the local government's wealth had positive and significant effect to the adoption of accrual-based Government Accounting Standard with value p-value of 0.007 <0.050. Hypothesis test results supported the hypothesis. The local government's wealth can be seen from PAD. Patrick (2007) stated that the current government or the central government provided these funds to the regions. In other that, the central government and provincial governments had to give accountability of the funds requested. The local government provided such accountability media through financial reports with the adoption of accrual-based Government Accounting Standard. Government Regulation 71 of 2010 stated that the purpose of financial reporting was to present information useful for decision making and to demonstrate the accountability of the reporting entity to the resources entrusted to it.
The 12 th Hypothesis test indicated that the local government's wealth had positive and significant effect to the financial performance of local government with value p-value of 0.202> 0.050. Hypothesis test results did not support the hypothesis of this study. The results of this study were not in line with research from Cohen S. (2008a) which stated that the assessment of financial performance using traditional financial ratios is influenced by macroeconomic factors, namely Gross Domestic Product, population, real estate value, tourist development, and government capital. However, (Cohen S., 2008a)reiterated that this financial ratio analysis still has drawbacks when used for cross-sectional comparisons. This is because macroeconomic factors of each region have its own characteristics which then will affect the financial performance of each local government. Therefore, the wealth of the region in this study has no effect on the financial performance of the local government because this research exists cross sectional analysis.
The 13 th hypothesis test indicated that the local government's expense had negative and no significant effect to the adoption of accrual-based Government Accounting Standard with p-value of 0.866> 0.05. Hypothesis test results did not support the hypothesis. The results of this hypothesis were inconsistent with Patrick (2007) which stated that local government with debt financing levels require a higher level of disclosure as a form of accountability to investors and creditors. The results of this study were not in line with research from Cohen S. (2008a) which stated that the assessment of financial performance using traditional financial ratios was influenced by macroeconomic factors, namely Gross Domestic Product, population, real estate value, tourist development, and government capital. However, Cohen S. (2008a) reiterated that this financial ratio analysis still had drawbacks when used for crosssectional comparisons. Macroeconomic factors of each region had its own characteristics which then will affect the financial performance of each local government. Therefore, the wealth of the region in this study has no effect on the financial performance of the local government because this research exists cross sectional analysis.
The 14 th hypothesis test indicated that the local government's expense of local government had negative and significant effect to the financial performance of local government with value p-value 0.003> 0.05. Hypothesis test results support the hypothesis. This means, the higher the operational cost owned by the local government, the resulting suplus on assets will also be smaller. The local government's expense was government liabilities recognized as net assets deductions. The largest operating expense was owned by Semarang city in 2015 and 2016. Semarang City also had the largest total assets in 2015 and 2016 so that it takes the operational cost to manage the large amount of assets. Therefore, the surplus generated by the Semarang city also small due to lower operating income and higher operating expenses to manage a large amount of assets.

Conclusion
Eight hypotheses are supported and eight hypotheses are not supported. The hypothesis supported is the educational background of the leader has a positive influence on the financial performance of the local government, experience using the accrual basis has a positive effect on the financial performance of the local government, the leadership of educational background has a positive effect on the financial performance of the local government, the adoption of accrual-based Government Accounting Standard mediates the influence of the leadership education on local government financial performance, the adoption of accrual-based Government Accounting Standard mediates the effect of experience using the accrual basis on local government financial performance, wealth has a positive effect on the adoption of accrual-based Government Accounting Standard, and the adoption of accrual-based Government Accounting Standard mediates the influence of local government's wealth on the local government's financial performance .
Based on the results of the study, all local governments in DIY have only implemented accrual-based Government Accounting Standard in 2015. All local government in Central Java started using accrual-based Government Accounting Standard also started in 2015. However, the Temanggung local government has started using accrual-based Government Accounting Standard in 2014. In addition, the percentage of accrual-based Government Accounting Standard rates in 2016 is higher than in 2016. This could indicate that experience and training in using accrual-based Government Accounting Standard must be increased so that errors related to recording and disclosure of accrual basis can be continuously improved by the local government and can be used appropriately.
Some limitations in this study, the data in this study are very limited, which is limited to provinces in DIY and Central Java in 2015 and 2016. Therefore, the results of this study cannot be generalized. The sample should use more samples so that the results of the study are more accurate, for example by using local government data throughout Indonesia. If further research uses samples from all local government in Indonesia, the results of the study can be generalized. This research tends to use cross section data. Measurement of financial performance using traditional financial ratios will be more appropriate if the data used is time series. If the measurement of financial performance uses traditional financial ratios, then further research can use time series data based on local governments that have almost the same characteristics The adoption of accrual-based Government Accounting Standard uses an accounting index whose valuation is only based on dichotomous values, namely 1 and 0. So, the level of assessment of the adoption of accrual-based Government Accounting Standard still does not show the percentage of accurate application. In addition, the variable influence of experience using the accrual basis on the adoption of accrual-based Government Accounting Standard can only be used for the years when the adoption of accrual-based Government Accounting Standard is applied. When all regions already have longer experience, then the possibility of using an accrual basis will not affect the adoption of accrual-based Government Accounting Standard. Measurement of the adoption of accrual-based Government Accounting Standard will be more accurate if using a research questionnaire to better understand the direct practice of the adoption of accrual-based Government Accounting Standard in the field or directly researched by people who are experts and experienced in examining local government financial reports.