The effect of BPK audit result on financial performance

Article history: Received: July 29, 2020 Received in revised format: July 3


Introduction
Indonesia has been included in the autonomous era where local autonomy provided the policies contained in Law No.22 of 1999 and then revised to Law No. 23 of 2014 about local government. Changes in the management of state finances make people aware that their responsibilities are in the form of aspirations that are useful in improving the management of state and local finances contained in government agencies or local government (Nasution, 2018). To do that, there are all processes, namely through planning, budgeting, budget execution, to financial accountability in the form of financial reports and finally audited (Juweny, 2016). Therefore, to get good governance must refer to PP No. 105 of 2000 which in managing local finance must done orderly, in accordance with the PP namely: influence, efficient and transparent. Therefore, the Government must make a financial report called LKPD (Local Government Financial Report) of course it must be in accordance with applicable law. BPK in Sumatra is divided in each province respectively. In each province there are LKPD (Local Government Financial Reports) which provide a lot of information. One of the information about opinions, audit findings, financial performance of local government which of course has been audited by the BPK. Each of the BPK audit results fluctuated from year to year. Seen from Table 1. From the data in Table 1, it appears that the results of the BPK audit in the form of opinion are already good in some provinces of Sumatra but the results do not look good in some other regions.  Masdiantini and Erawati (2016), Purwanto and Harto (2017), report that audit findings did not affect the financial performance of local government. This result is also supported by Agustiawan (2018) in which the results of audit opinions and findings did not influence financial performance. Halimah's research (2017) found that opinions audited by BPK did not influence the financial performance. On the contrary, Handayani (2019) reported that the audit findings did not have any positive influence on financial performance. The conclusions of Lukman's (2019) and Iqbal (2018) did not report any significant relationship between audit findings and financial performance. Others also show the same results that audit opinion has a negative effect on financial performance. With the above data and the inconsistency of the results of previous studies, therefore, this paper is interested in studying the influence of the results of the audit on financial performance in the districts / city government in Sumatra.

The Agency Theory
The Agency theory is the basic theory in this research. The connection of the impact of BPK audit results with financial performance explains in this agency theory. Described by Jesen and Meckling (1976) this agency theory has a relationship as "Agency relationship as a contract under which one or more people (the principals) engage another person (the agent) to perform some service on their behalf which involves delegating some decision making authority to the agent". The linkage to the agency is where there is an agreement between the individual or group (principal) giving responsibility to the other person (agent) in the form of services to issue the best decision for the principal. Sourced by this agency theory explains that there is a relationship between the government (agent) and the community (principal) due to the contract (contract) established by the community. Not only is it limited to assigning tasks, the principal also continues to monitor and hold accountability for the resources managed in the form of financial statements. The problem arises from differences in ownership owned by the principal and managed by agents (Sudarsana & Raharjo, 2013). With this mental agent it is possible to achieve some benefits in principal, but in other parts this agent also optimizes its survival. Government is most likely to get information about the resources available in the area in the form of APBD from the public, this is what is known as information asymmetry and problems as agency occur.
Because LAKIP / LKPD contains information about performance results, agency problems can be overcome because the community (principal) can directly measure and monitor government performance. From the explanation above, the government as the principal is obliged to make financial reports because agents (the public) need information from the report.

Findings Audit
Findings are a systematic inspection process carried out to find out how actual implementation is determined (Afrian, 2016). Financial performance at the local government will decrease if the audit findings increase (Agustiawan, 2018). The number of findings obtained by the BPK is a grouping of many or at least a finding with reference to the average value of the findings that are available about SPI (Afrian, 2016).

Opinion Audit An
Examination of financial statements conducted by the auditors will result in the form of an audit opinion and the better the opinion obtained, the better the financial performance achieved at the local government (Masdiantini & Erawati, 2016).

Financial Performance of Local Governments
Financial performance according to Law No. 32 of 2004 is all rights and responsibilities of the region and it can be of cash and items that can be applied as local right of ownership associated with the implementation of these rights and obligations. Financial performance can be reviewed through analysis of local government financial statements (Islami, 2020). In this paper, local government financial performance is measured using the influenceiveness ratio with the following percentage ranges: below 60% no influencive; 60% -80% less influencive; 80% -90% are quite influencive; 90% -100% influencive and 100% and above are very influencive. According to Retnowati (2016) and Masdiantini (2016), the following formula is used for measuring the financial performance of local governments. ℎ =

Research Model
T test is performed for hypothesis testing and random effect model (REM). In random effect model (REM), Equations are formed into models, the equation is: where: X 1 = Audit Opinion, X 2 = Audit findings, Y= Financial Performance

Hypothesis
The hypotheses of this paper are as follows, H 1 : Audits opinions have positive effects on financial performance. H 2 : Audits findings have negative effects on financial performance.

Research Methodology
This paper aims to search for about the effects of audits opinions and findings opinion on financial performance of local government in Sumatera, Indonesia. This paper uses quantitative research approach and some secondary data were obtained from document analysis. The document analysis consists of data from audit opinion and Audit findings. Data of this paper were collected from 130 financial performance of districts / city local government (LKPD) in Sumatera, Indonesia. Eviews 9.0 becomes a statistical procedure for analyzing data in this paper.

Sampling Technique
Samples in this paper are financial performance of districts / city local government (LKPD) in Sumatra, Indonesia taking into consideration that local government was audited by BPK, and do not have LKPDs in complete 3 years from 2016 to 2018. To determine the sample size there are certain criteria. From these criteria it can be seen that from 492 populations, the sample included only 390.

The Model Feasibility Test Method
The feasibility test of the model uses 3 approaches, namely: In this paper, the multicollinearity and the heteroscedasticity tests are executed.

Results
Findings of the study consists of the model test, classic assumption test, and the hypothesis test. The test model is carried out through three approaches, namely common influence or polled least square (PLS), fixed effect model (FEM) and common effect model (CEM). Some of the tests in this study will be used to determine the right estimation model to use, while the tests are as follows Chow test, Hausman test, and Lagrange multiplier test. Classical assumption test is a multicollinearity test and a heteroscedasticity test. Hypothesis test discusses the F test and t test.  Table 4 show the value of the probability cross section F = 0.0001 <0.05. F-calculated value 1.711584 <F Table of 3.019042 and the model of common effect is used in comparison with the fixed influence for estimating panel data.  Table 5 it can be seen that the probability value is 0.4926 or greater than the alpha 0.05, then the random effect model is better than the fixed. Lastly Lagrange multiplier given in Table 6 is a test that will decide the random effect model or common effect model corresponding to be used in panel data regression. It can be seen that the value Lagrange multiplier that the significance value is equal to 0.0000 <0.05 which means the right estimation model is the random effect model.   Table 9 shows the results of the Heteroscedasticity test.  According to Table 9, the.heteroscedasticity shows that all independent variables are statistically significant which means there is no heteroscedasticity. Next, we present the results of some other statistics associated with the proposed study in Table 10.  Table 10 shows that Durbin-Watson statistics is equal to 1.588710 which means there is no collinearity between the residuals.

classic assumption test
In summary, we can conclude that there is no relationship between independent variables; namely, Audit Opinion and Audit findings, and Financial Performance.

Conclusion
In this paper, we have presented an empirical investigation to study the effects of audit opinion and audit findings on government financial performance. The study collected the necessary information from different regions of Indonesian governmental firms. The results of the regression analysis have indicated that none of two independent variables had any meaningful effect on financial performance. The results of this survey are consistent with the results of Lukman's (2019) and Iqbal (2018) where they reported no significant relationships. Masdiantini and Erawati (2016), Purwanto and Harto (2017) also reported that audit findings did not affect the financial performance of local government. The survey has some limitation which is associated with the region and limitation on the number of sample size. Future studies could extend the independent variables and perform more detailed results.