“Key elements of local government transparency in new public governance”

Transparency is a challenge for public administration reforms for now and beyond towards modern and better governance. The economic crisis that occurred has led to a large erosion of public confidence in the effectiveness of government financial manage- ment. This study aims to analyze the key elements of local government transparency in new public governance. The research data was obtained through financial reports on local government websites. The sample selection technique used purposive judgment sampling and yielded a total of 262 local governments in Indonesia. The analyti- cal method used is logistic regression analysis. The results showed that key elements, reflected by the local government size and local financial independence ratio, had a significant influence on the implementation of internet financial reporting, while audit opinions do not have a significant effect. This is due to changes in public perception of the quality of audit opinions provided by the Audit Board. This study emphasizes that local governments that receive a fair or unfair audit opinion cannot prove with cer- tainty that the local government has performed good or poor financial management. Audit opinions obtained by local governments do not always facilitate the publication of financial statements on their official websites. corruption. This is in line with the results of a survey conducted by Transparency International Indonesia. The lack of transparency is often associated with corruption in the government sector. Sharman and Chaikin (2009) show that corruption is an important problem in most developing countries and is the biggest hindrance to economic development. Ellis and Fender (2006) show that economic transparency, which results in a decrease in the level of corruption, is part of the output over time, while less transparent economies face an increase in corruption. on the Perception Survey it was revealed that


INTRODUCTION
Transparency, among other things, is interpreted as the extent to which an organization provides information about its internal work, decision making processes and available procedures (Pina et al., 2010). According to Indonesian Corruption Watch (Wiwoho, 2014), during the first semester of 2014, the trend of corruption defendants was the highest among local government officials or employees, out of 261 accused of corruption in 2014, there were 101 local government officials who were proven to be corruptors. The number of local government officials as corruptors increased from the previous two years. In 2012, Local original income ICW report found 48 local government officials who committed corruption, and in 2013, the number increased to 60 local government officials accused of corruption. This is in line with the results of a survey conducted by Transparency International Indonesia. The lack of transparency is often associated with corruption in the government sector. Sharman and Chaikin (2009) show that corruption is an important problem in most developing countries and is the biggest hindrance to economic development. Ellis and Fender (2006) show that economic transparency, which results in a decrease in the level of corruption, is part of the output over time, while less transparent economies face an increase in corruption.
Based on the Indonesian Corruption Perception Index Survey Report issued by Transparency International Indonesia, it was revealed that in 2014 Indonesia's corruption perception index score only increased by 2 digits, from 32 to 34, and Indonesia's ranking increased from 114 in 2013 to 107 in 2014 among 175 countries. Indonesia is still far behind Singapore and Malaysia. This indicates that political corruption in Indonesia is still dominant and difficult to eradicate. It influences the emergence of a public confidence crisis in the government performance. Consequently, the government must strive to rebuild public trust by transparency increasing, especially in the local government finance management. Problems related to the implementation of e-governance have been highlighted by Chatterji (2018), neglect has made cities in India weak and lacked skilled human resources in the field of information technology and electronic governance.
The government must be positive and open to the public, without limiting the amount of information that discloses the government's performance. The publication of financial statements is a major element that can represent the accountability of financial statements (Adriana & Ritonga, 2018;Sargiacomo & Gomes, 2011;Yusuf & Jordan, 2015;Waheduzzaman, 2019). An open government strategy is one of the objectives of promoting transparency, which involves the publication of government data to support accountability and reuse for social or economic value. Public accountability is data disclosure that provides information needed by citizens and other public stakeholders to assess the appropriateness and effectiveness of government (Lourenço, 2013).
Some previous studies have actually tried to explore empirical evidence about the factors that influence the disclosure of local government financial statements on the internet. Recently, demands for transparency in the government system have increased. One factor that has led to increased demands for local financial transparency is the economic crisis, which has led to a large erosion of public confidence in government financial management performance. This public confidence crisis in the government is caused by poor financial management of local government, which is reflected in the number of local government officials who have been dragged into corruption cases. This study seeks to conduct a deeper test of several factors that are believed to affect the internet financial reporting in Indonesia's local government.

LITERATURE REVIEW
Disclosure of financial information via the internet is called Internet Financial Reporting, which is a mixture of internet multimedia capabilities and communicating it interactively (Warsidi et al., 2017;Styles & Tennyson, 2007). The internet can be used as a media related to the preparation and budget use or the publication of local government financial reports to increase transparency between the government and community. According to Pina et al. (2010) and Bertot et al. (2010), advances in information and communication technology are effectively seen as able to minimize costs to create transparency and reduce the corruption level in the government sector. The diminished transparency is felt along with the lack of trust in the government (Yusuf & Jordan, 2015).
Voluntary disclosure of local government financial reports via the internet is considered efficient and effective in increasing control of local government instruments for acts of corruption (Guillamón et al., 2011). Further development of the two-way e-government communication function is a reform strategy that will lead to service-oriented government (Jun et al., 2014). Martani et al. (2014) show that the size, dependence level on the central government, and local government welfare positively affect transparency level of financial information and performance of Indonesian local government. Fadah et al. (2017) show that a local government website has not been used optimally in developing financial reporting. Research by Adiputra et al. (2018) showed that the quality of financial reporting through audit opinion and the political environment positively and significantly affected transparency of local governments in Indonesia.
In general, the size of local government positively affects the amount and transfer of local government wealth. The greater amount and transfer of local government characterizes wealth that the greater of the local government size. Large local governments have more complex financial management. Financial reporting requires a lot of financial information to minimize information asymmetry. Aronmwan and Asiriuwa (2018), Jimoh and Okoye (2016), and Laswad (2005) show that the local government size will encourage the publication and disclosure of financial statements via the Internet, while Perez et al. (2014), Nosihana and Yaya (2016), and Stephen et al. (2017) reveal that the local government size has no influence on the financial reporting of the local government through the internet. Large-scale local governments require more oversight and accountability, which forces local governments to bear higher supervision and reporting costs. The internet is the most effective and satisfying media for local governments in publishing financial reports in the context of implementing transparency and accountability in financial management. Through the internet, local governments can provide financial information to the public quickly, in detail and at a low cost.
Audit opinion as one indicator of the quality of local governments' financial accountability is estimated to have an influence on the publication of local government financial reports via the internet in Indonesia. The performance of local governments can be assessed not only based on the financial reporting quality, but can also be evaluated broadly on the local government administration performance carried out by the central government through the Ministry of Home Affairs. Thus, local governments with better audit opinions are also expected to have a higher performance of local government administration. Some considerations for expressing audit opinions are deficiencies in the internal control system, non-compliance with regulations and laws, and non-compliance with Government Accounting Standards ( and Sutopo et al. (2017) show that an audit opinion is positively related to the performance of local government administration. Local governments that get a fair opinion will tend to publish financial reports via the Internet to inform about good and accountable financial management quality. Johnsen et al. (2001) revealed the efficiency of the audit in the regional government, which was assessed by auditors and an auditee in Finland and Norway. Although there are issues relating to the quality of performance audit reports, informants consider them to be useful and rational public management tools. Akrom and Firmansyah (2017) state that audit opinion weakens the effect of the local government's financial independence ratio on compliance with financial information disclosure on the website. Audit opinions cannot moderate the effect of local budget ratios absorption and effectiveness ratios on the compliance with the financial information disclosure on the website.
The independence level shows the region's ability, from local revenue sources to local operational expenses finance, in the implementation of regional development and public services. Local original revenue is the income of an autonomous region from the potential area and other funds that are approved based on law used by local government for progress and welfare of the autonomous regional community. The local own revenue ratio is often referred to the local financial independence ratio. According to Christiaens (2002), local government that has high local original revenue will show stakeholders that the local government has produced high performance. Local governments with poor performance will avoid voluntary disclosure and prefer to limit access to information for the public ( Puspita and Martani (2013) show that the local original income ratio does not affect the disclosure and the quality level of information on the local government website, whereas Annisa and Murtini (2018) show that the local financial independence ratio has a significant positive effect on the financial information availability on the original local revenue official website of the local government.

METHOD
The research design used in this study is quantitative research. This research was conducted by observing the official website of local government, especially the regency/city government in Indonesia. The study uses purposive judgment sampling techniques. The sample selection summary in this study is shown in Table 1. where LN_SIZE -Local government size, DOPI -Audit opinion, RKKD -Local financial independence ratio.  Table 3 shows that from the total sample, the local governments implementing IFLGR on the official website amounted to 128 or only 48.85 percent of the total sample. While the remaining 134 (51.15 percent) local governments did not implement IFLGR on their official websites. This shows that the implementation level of IFLGR in Indonesia is still low. Local governments were unable to make the most of the local government website to deliver its financial information.    Table 5 shows that the correlation value between independent variables does not exceed 0.70, which indicates no strong multicollinearity symptoms among the independent variables in the study, so the regression model is good. The Chi-Square Goodness-of-Fit test with several measurements is used to measure the feasibility of a regression.

RESULTS
The overall model is assessed by comparing the value between -2 Log Likehood (-2LL) at the beginning (Block Number = 0), where the model only includes constants, with the value of -2 Log Likehood (-2LL) at the end (Block Number = 1 ), where the model includes constants and free variables. Nagelkerke's R Square test is used to show the magnitude of the coefficient value of the logistic regression model determination. The value of the determination coefficient for logit regression can be seen in Table 7.  Table 8 shows the results of the Hosmer and Lemeshow test.

Table 8. Hosmer and Lemeshow's test results
Step Chi-square df Sig. 1 12.714 8 .122 Table 8 shows that the significance value of this study is 0.122. Based on these results, it can be concluded that the model can explain empirical data, since the significance value exceeds 0.05. In other words, H 0 is accepted, that is, a binary logistic regression model can be accepted as it is in accordance with the observational data.
Classification table is used to see how well the prediction of the research model is used in this study. For a perfect model, the overall percentage will be 100 percent. The predictive power of the regression model to forecast the like lihood of the dependent variable is expressed in percent. Table 9 shows that the overall percentage that correctly predicts the model has a good enough value of 61.5 percent.  Table 9 shows that the predictive power of the regression model in predicting the likelihood of a local governments implementing IFLGR is 49.2 percent. This shows that using the regression model, there are 63 local governments predicted to implement IFLGR out of selected 262 local governments. The predictive power of the regression model in forecasting the likelihood of local governments that did not implement IFLGR is 73.1 percent. This means that with the regression model, there are 98 local governments predicted to unimplement IFLGR. Based on Table 10, it can be seen that the significant probability value for the size of the Regional Government is 0.039. This value is below the probability value α = 5 percent. Thus, it can be concluded that H 1 1 was accepted and H 0 1 was rejected, which means that the size of the Local Government has a significant effect on the Implementation of Internet Financial Local Government Reporting in Indonesia. The variable coefficient of the Local Government Size is 0.545, which means that each one-unit increase will result in an increase in the level of Implementation of the Internet Financial Local Government Reporting by 0.545 units. To analyze the second hypothesis and to find out a partial effect of the Audit Opinion on the Internet Financial Local Government Reporting Implementation in Indonesia, the following hypotheses are formulated:  Table 10 shows that the significant probability value for the audit opinion is 0.308. This value is above the probability value α = 5 percent. Then, it can be concluded that H 1 2 is rejected and H 0 2 is accepted, which means that audit opinion does not affect the Implementation of Internet Financial Local Government Reporting in Indonesia. The variable coefficient of the Regional Government Size is -0.570, which means that every increase of one unit will result in a decrease in the level of the Internet Financial Local Government Reporting Implementation by 0.570 units.
To analyze the third hypothesis, which should know the partial influence of the Local Financial Independence Ratio on the Implementation of Internet Financial Local Government Reporting in Indonesia, the hypotheses to be tested are formulated as follows: The greater the total assets owned by local governments, the greater the resources that are allocated by local governments for developing their sites. Styles and Tennyson (2007) state that the higher the level of the local government wealth, the higher the monitoring by the community, and the higher the demand for information provided on the website to measure the local government performance. Greater community demands for the performance transparency implementation, especially financial performance by the local government, force largescale local governments bear higher supervision costs to meet the transparency demands from the public. In addition, large-scale local governments need more financial information that must be reported to minimize information asymmetry between public and government. This leads to the fact that large local governments have to bear high reporting costs.
The results of this study are in line with the Stakeholder Theory, which states that the more powerful the stakeholders (the community), the greater the government's adaptation efforts, so that it will later receive full support from the community. The critical attitude held by the public towards the management of government finances has an impact on the government's increasingly large efforts to improve transparency by utilizing cost effective reporting media as its financial reporting media.

The effect of the audit opinion on the Internet financial reporting implementation
Fair audit opinion obtained by local governments does not necessarily encourage the local governments to implement internet financial local government reporting on each of their official websites. The reasonable audit opinion obtained by the local government cannot prove with certainty that the local government has carried out good financial management. A non-fair opinion received by the local government does not mean that the credibility and accountability of the local government are bad.
In addition, changing public perception about audit opinions has resulted in audit opinions not directly influencing local governments to publish financial reports via the Internet on the official websites of each local government. Changes in public perceptions about the audit opinion occurred due to the rise of cases of local governments that intentionally made a bribe to get a fair audit opinion from the Audit Board. The number of cases shows the public that a fair audit opinion is not a guarantee that the local government is free from corruption. Fairness does not mean the truth of a transaction. The higher ratio of local financial independence indicates the better performance and financial management of the local government. Local governments with a high level of the independence ratio will strive to show their stakeholders that they have good performance that encourages local governments to publish their financial reports on the internet. According to the results of this study, the size of the local financial independence ratio represented by a comparison of values with the total local government revenue has a significant positive effect on the implementation of internet financial local government reporting in Indonesia.
The results of the study are in line with the signaling theory, which explains how signals of agents' success or failure are conveyed to the principal. The good financial performance of the government reflected in a high level of the independence ratio will provide good news, which is a positive signal for parties outside the government, such as the public. Local governments with good quality management tend not to restrict access to information and prefer to disclose more information and use systems that can improve the quality of local governments by providing financial information on their official websites.

CONCLUSION
Thus, it is hoped that the local government can adopt the most effective and efficient reporting method. The internet is considered as the most effective and efficient reporting medium in terms of cost and time, so that it can provide a feedback value for the local government itself. By developing mass media on local government performance, which relates to the design and optimization of websites owned by each local government, transparency demands for the public or community can also be met. The implications of the results of this study for large-scale local governments are as follows -they will be able to develop their sites as reporting media, since they are considered more efficient and effective when viewed in terms of cost and time compared to traditional media, so that all resources allocated by local governments to develop regional sites into media in performance reporting provide benefits for local government, can reduce information asymmetry and meet the transparency demands from the public. Audit opinion does not affect the implementation of the internet financial local government reporting in Indonesia. This is due to changes in public perception of the quality of audit opinions provided by Audit Board (BPK). A local government that receives a reasonable audit opinion cannot prove with certainty that the local government has performed good financial management, while a local government that receives a non-fair audit opinion does not necessarily indicate poor financial management of local government. A higher level of regional financial independence indicates that local government performance is also improving, so local governments with the high regional financial independence ratio will tend to demonstrate to stakeholders that they are interested in publishing their financial statements on the Internet.